+Local Development and Building MLOps Packages
After you finish the local developing/debugging of the FedML project using FedML library (e.g., successfully run the example https://doc.fedml.ai/cross-silo/examples/mqtt_s3_fedavg_mnist_lr_example.html),
you can now deploy it into the real-world edge-cloud system.
Next, we use the cross-silo project example at
@@ -553,7 +553,7 @@
5. Experimental Tracking via Simplified Project Management
- How Does FedML MLOps Platform Work?
-- Local Development (Simulation) and Building MLOps Packages
+- Local Development and Building MLOps Packages
- 1. Create Application and Upload Local Packages
- 2. Install FedML Agent: fedml login $account_id
- 3. Invite Collaborators, Create a Group and a Project
diff --git a/doc/en/_build/html/searchindex.js b/doc/en/_build/html/searchindex.js
index ceeead762c..09761c68f9 100644
--- a/doc/en/_build/html/searchindex.js
+++ b/doc/en/_build/html/searchindex.js
@@ -1 +1 @@
-Search.setIndex({docnames:["README","api_doc/api-core","api_doc/api-fedml-api","benchmark/index","cross-device/api","cross-device/examples","cross-device/examples/mqtt_s3_fedavg_mnist_lr_example","cross-device/user_guide","cross-silo/api","cross-silo/examples","cross-silo/examples/mqtt_s3_fedavg_hierarchical_mnist_lr_example","cross-silo/examples/mqtt_s3_fedavg_mnist_lr_example","cross-silo/user_guide","distributed/api","distributed/examples","distributed/user_guide","index","mlops/api","mlops/examples/mlops_live_demo","mlops/mlops_user_guide","mlops/t-mlops","mlops/user_guide","resource/community","resource/papers","resource/slides","resource/video","simulation/api","simulation/benchmark/BENCHMARK_MPI","simulation/benchmark/BENCHMARK_simulation","simulation/examples","simulation/examples/fedavg-mnist-examples","simulation/examples/mpi_torch_fedavg_mnist_lr_example","simulation/examples/sp_fedavg_mnist_lr_example","simulation/user_guide","simulation/user_guide/algorithm-reference-implementation","simulation/user_guide/data_loader_customization","simulation/user_guide/datasets-and-models","starter/concepts","starter/ecosystem","starter/examples","starter/faq","starter/get_started","starter/install/windows","starter/installation","starter/mission","starter/mlops_video","starter/oss_code_architecture","starter/overview"],envversion:{"sphinx.domains.c":2,"sphinx.domains.changeset":1,"sphinx.domains.citation":1,"sphinx.domains.cpp":4,"sphinx.domains.index":1,"sphinx.domains.javascript":2,"sphinx.domains.math":2,"sphinx.domains.python":3,"sphinx.domains.rst":2,"sphinx.domains.std":2,sphinx:56},filenames:["README.md","api_doc/api-core.md","api_doc/api-fedml-api.md","benchmark/index.md","cross-device/api.md","cross-device/examples.md","cross-device/examples/mqtt_s3_fedavg_mnist_lr_example.md","cross-device/user_guide.md","cross-silo/api.md","cross-silo/examples.md","cross-silo/examples/mqtt_s3_fedavg_hierarchical_mnist_lr_example.md","cross-silo/examples/mqtt_s3_fedavg_mnist_lr_example.md","cross-silo/user_guide.md","distributed/api.md","distributed/examples.md","distributed/user_guide.md","index.rst","mlops/api.md","mlops/examples/mlops_live_demo.md","mlops/mlops_user_guide.md","mlops/t-mlops.rst","mlops/user_guide.md","resource/community.md","resource/papers.md","resource/slides.md","resource/video.rst","simulation/api.md","simulation/benchmark/BENCHMARK_MPI.md","simulation/benchmark/BENCHMARK_simulation.md","simulation/examples.md","simulation/examples/fedavg-mnist-examples.md","simulation/examples/mpi_torch_fedavg_mnist_lr_example.md","simulation/examples/sp_fedavg_mnist_lr_example.md","simulation/user_guide.md","simulation/user_guide/algorithm-reference-implementation.md","simulation/user_guide/data_loader_customization.md","simulation/user_guide/datasets-and-models.md","starter/concepts.md","starter/ecosystem.md","starter/examples.md","starter/faq.md","starter/get_started.md","starter/install/windows.md","starter/installation.md","starter/mission.md","starter/mlops_video.rst","starter/oss_code_architecture.md","starter/overview.md"],objects:{},objnames:{},objtypes:{},terms:{"0":[1,6,10,11,21,27,28,31,32,34,41,43],"001":[10,11,27,32,34,41],"003":[27,32],"01":[27,41],"014680":32,"018868":11,"02":[6,31],"03":[11,27,28,32,34,41],"04":[6,10,11,31,32,43],"046":32,"051":32,"053":32,"053363":11,"054":31,"055":31,"057":32,"059":32,"060":32,"062":32,"064":32,"067":32,"068":32,"07":10,"070":32,"07145":38,"071570":32,"07976":38,"08815":38,"093762":32,"0x000001cf758d8f88":32,"0x9801a7a55e85":19,"0xb6ff42da6a7":21,"1":[1,10,11,16,23,27,28,34,39,41,43,44,46],"10":[4,6,10,11,23,27,31,32,34,41,42,43,47],"100":[10,11,27,31,32,33,36],"1000":[10,11,27,28,32,34,41],"104":31,"105":19,"107":11,"11":[23,42,43],"11066":38,"115":10,"116":10,"118":31,"12":[23,27,32],"1200":27,"124":[10,32],"127":[21,43],"128":[6,11],"128363870":32,"129":[6,10],"13":23,"130":32,"132":[10,32],"132013":32,"13518":27,"137":[10,11],"14":[23,32,41],"140":19,"14285715":1,"145":11,"15":[10,23,32],"1500":27,"155":10,"156245":32,"16":[10,23,27],"160":10,"164":[6,10],"165":10,"1651030718":11,"16666667":1,"168":10,"17":32,"172":10,"18":[27,28,29,32,41],"186":10,"189":6,"19":[27,28,32],"191":10,"1912692754254417":10,"192":10,"197":[10,32],"198":32,"199":[11,32],"1ba9":10,"1canbw":28,"1d4b":11,"2":[1,10,11,16,23,27,28,33,36,39,41,43,44,46],"20":[10,11,23,27,28,31,32],"200":[6,22,27,28,32,34,41],"2000":28,"2007":27,"201":6,"2018":23,"2019":23,"2020":[23,25,33,34,41,43,46],"2021":[23,33,34,38],"2022":[6,10,11,23,31,32,38,41,46],"204":10,"21":[23,27,32,38],"2104":38,"2106":38,"2111":38,"212":10,"212439":32,"215":10,"22":[6,23,31,41],"221":32,"222175":32,"224":10,"225":11,"23":27,"235":10,"237719":32,"240":11,"244":10,"246":10,"25":[10,27,32],"250":10,"253":10,"254":10,"257":10,"26":[1,31,32,43],"261":10,"262":11,"263015":10,"266":21,"269":10,"27":[6,11,27],"271503":10,"276":32,"276260":10,"28":[6,10,11,31,32,41],"284570":10,"284942":10,"288988":10,"29410":10,"295648":32,"296":[10,11],"296894":10,"29th":23,"2al5q5mi":28,"3":[1,10,11,16,23,27,28,33,41,42,43,44,46],"30":[10,27,31,32,33,41],"300":6,"301717":10,"302":21,"306":11,"306870":10,"308":[10,32],"31":[6,10,41],"317":32,"32":[6,27,43],"33":[10,31],"33333334":1,"337":10,"34":[28,32],"3400":27,"342477":27,"36":38,"369121":32,"37611896730669436":10,"3785d45dce8911ecbae4ea89f38f9f29":42,"38":[11,32],"3aponqml":28,"3lv4gmpz":28,"4":[1,16,23,27,43],"40":[10,11,27,32],"4000":[27,28],"41":32,"41b8":11,"42":32,"424e":10,"436":32,"44":[10,27,28,32],"444":32,"445c":11,"45":[11,32],"47":[10,11],"479":32,"48":[6,10],"485":32,"49":[6,10,11,27,31],"4ae765659471":11,"4b6d":10,"4btyrt0u":28,"4d9d":10,"5":[1,10,11,16,19,23,27,28,31,32,41,43],"50":[10,11,23,27,32,41,47],"500":27,"506":32,"50th":[10,11],"51":11,"516":32,"52":32,"53":[6,27,28,31,43],"54":27,"54615336":42,"55":[10,27],"56":[11,27,31],"57":[10,27,28,32,41],"5741":11,"58":11,"5836178064346313":6,"59":41,"591":11,"6":[1,19,23,27,43],"60":[27,32],"600":[33,36],"602":11,"62":32,"63":[11,31,32],"64":[4,27,42],"64g":43,"65535":43,"661":32,"67":[11,31],"68":27,"685":32,"6be8":10,"7":[1,21,23,27,28,31,38,41,42,43],"70":[27,32],"71":[27,31],"711":[6,32],"715":27,"7155714841722886":11,"7156":11,"717948717948718":11,"73":27,"7344120322737784":32,"73456031979222":32,"74":11,"75":[10,27,32],"750":32,"76":6,"77":6,"775815":31,"777":43,"784":32,"79":27,"794":32,"7pf2c9r2":28,"8":[1,23,27,28,41,43],"80":[10,28,32],"803107":11,"81":[11,28,32],"817":32,"818":32,"8188848188848189":32,"8190029839128179":32,"82":27,"83":6,"839799":31,"84":[27,28],"845405":31,"86":27,"860":32,"861":32,"866":32,"867":32,"867b":10,"87":27,"88":[6,32],"887":32,"888":32,"891":32,"8972983557921448":11,"8997":11,"8997359397010631":11,"8b7e880db4c5":10,"9":[23,27,28,31,43],"90":17,"905":32,"91":[6,27],"920599":31,"9205eea272f2":11,"925":32,"93":27,"930":11,"93setzgzmyi":46,"94":10,"942":32,"95":27,"954":32,"959164":32,"96":32,"966263":32,"973":32,"984373":11,"986":32,"9c71":10,"\u03b1":27,"\u03b2":27,"abstract":1,"break":23,"case":[29,30,31,32,33,36,41,43,44],"class":[1,10,11,31,32,33,35,36],"default":[19,41,43,46],"do":[1,10,11,18,19,20,21,31,33],"final":43,"function":[1,19,31,32],"import":[1,4,8,10,11,23,26,31,32,34,41,43],"int":1,"new":[2,4,8,19,21,26,32,33],"null":[],"public":[16,33,36],"return":[10,11,31,32,35],"super":[10,11,23,31,32],"switch":[6,41],"true":[4,10,19,21,32],"try":[4,8,18,20,26,29,32,33,41],"var":43,"while":[1,8,10,26,28,38,41],A:[16,19,23,35,46],As:[1,9,19,20,31,32],At:[10,11,16,32],Being:23,By:[19,25,41],For:[1,4,6,8,10,11,26,27,31,32,33,34,35,36,41,43],IN:1,If:[6,10,16,19,21,33,41,43],In:[1,10,21,29,30,31,32,33,38,41,43,44,46],It:[1,16,18,20,41],Its:[16,41,46],One:[25,41],Or:19,Such:[2,18,19,20,21,41],The:[1,6,8,10,11,18,19,20,21,28,31,32,34,35,38,41,42,46,47],Their:36,Then:[6,21,32],These:38,To:[4,6,8,10,11,26,33,34,38,41],With:[1,2,38,41],__finish:[10,31],__init__:[1,10,11,31,32,41],__main__:[4,8,10,11,26,31,32,34,41],__name__:[4,8,10,11,26,31,32,34,41],_notifi:10,_on_disconnect:10,_on_message_impl:10,a1xfouratnfc9dfqu:[22,41],a925:11,aaai:[23,38],abc:1,abhishek:[41,46],abil:[19,21],abl:[10,19,21],about:[16,19,38],abov:[1,6,18,19,20,21,31,41,43],absolut:21,abstractmethod:1,academia:[41,44],acc:[6,27],acceler:[10,23,29,30,41,44,46],accept:[19,21],access:10,accord:[1,19,32],account:[6,21,22,41,44,46],account_id:[16,19],accur:23,accuraci:[6,11,23,27,28,38],achiev:[10,23],acm:[23,38],across:16,activ:[32,42,43],actual:38,ad:1,adam:38,adapt:[23,27,38],adavantag:10,add:[4,8,19,21,22,26,41],addit:[1,33,43],address:10,adjust:23,admin0000:[18,20],admin:[18,20],adopt:2,advanc:[19,21,23,38,41,43,46],advantag:38,afford:38,after:[1,6,19,21,41,43,46],agent:[10,11,16,19],aggreg:[1,6,9,10,21,23,41],aggregator_dist_adapt:10,agnost:23,ai:[1,4,6,8,16,18,19,20,21,26,27,28,31,32,33,35,36,40,41,43,44,46,47],aim:[38,41,46],aistat:23,alexliang:19,alexqzliang:19,alg:27,algorithm:[1,2,16,19,29,32,36,38,41,46],all:[1,9,10,16,19,21,31,32,38,41,43,46],all_reduc:1,alloc:27,allow:[1,10,38],allreduc:16,almost:31,alogrithm:27,alon:[31,32],along:38,alpha:27,also:[1,2,6,8,11,16,17,19,21,23,26,27,31,32,41,46],although:1,alwai:[4,8,23,26],among:[1,2],amount:38,an:[1,9,21,23,31,38,41,43,46],anaconda3:41,analysi:38,analyt:[16,23],android:[4,5,16,41,44],android_dir:6,angel:[18,20],ani:[1,16,18,19,20,21,35,41,42,43,44,46],annavaram:[41,46],anomali:38,anoth:[1,2,19],answer:[33,36,38],anywher:[16,18,20,21,41,44,46],api:[16,34,44,46],app:[4,27,28],appendix:27,appli:[16,31,32,33,38,41],applic:[1,2,6,16,33,38,41],appreci:[19,21],approach:[23,38],approxif:23,approxim:23,apr:[6,11,31],apt:43,aq:43,ar:[1,9,10,11,16,19,21,28,32,33,35,36,38,40,41,43,46],arbitrari:1,architectur:[2,10,16,23,43,47],area:23,arg:[4,8,10,11,26,31,32,34,35,36,41],argument:[1,4,8,19,21,26,31],arm64v8:43,articl:[23,41,46],arxiv:[23,27,38],ask:[22,40,41],aslo:10,assign:31,assist:[19,21],assum:33,asymmetr:[1,23],asymmetrictopologymanag:1,asymml:23,async_html:0,asynchron:23,attack:[23,36,38],attain:38,attribut:10,author:[41,46],auto:19,autoct:23,autom:[16,23],automat:[10,11,41],automl:[27,28],auxiliari:1,avail:[10,23,31,32,38],averag:[23,39],avestimehr:[25,41,46],avoid:[2,31],await_sync_process_group:10,awar:23,award:[23,25,41,46],b3d7:11,b6c6:10,b6ebbab3ce8a11ecb8af70df029276d3:42,back:[1,16,23,41,47],backdoor:36,backend:[1,10,11,16,31,32,41,46],background:1,bandwidth:[23,29,30,41],bank:[9,41],bar:6,barrier:23,bart:38,base:[1,2,16,23,32,33,34,38,39,41,42,43,44,46],baselin:33,basetopologymanag:1,bash:[6,10,11,31,43],basic:[31,43],basil:23,batch:[6,23,27],batch_siz:[4,10,11,27,28,31,32,34,35,41],becaus:[1,31],beehiv:[6,44],been:[33,38,41,42,46],befor:[1,6,19,21,33],behavior:1,behind:38,believ:38,below:[1,9,10,11,19,21,31,32,33,34,43],benchmark:[16,23,41,46],benefit:[2,23,33,38],bert:38,besid:23,best:[4,8,23,25,26,38,41,46],better:[38,41,44],between:[10,23,33,38],bi:36,bia:32,bin:[6,10,11,31,43],binari:6,bind:21,bit:42,blob:[4,8,17,26,35],blockchain:23,blue:1,boostrap:43,bootstrap:[10,43],both:[38,41,44],bottom:6,bound:[19,21],box:[1,47],brand:6,bridg:38,bring:33,broad:16,broadcast:10,bs:27,buffer:23,build:[1,16,18,20,38,41,42,44],built:[1,2,33,36,38,41,46],busi:33,button:21,byzantin:23,c289:11,c:[0,16,27,42,43],c_opt:27,c_p:27,ca9194865e31:10,call:[1,10,11,32],can:[1,2,6,9,10,17,18,19,20,21,31,32,33,35,36,38,41,43,46],cannot:[1,2],capabl:[16,41,46],capac:38,care:43,career:41,caus:31,cd:[21,43],central:[1,16,18,20,38,41,46],cf:[10,11,17,21,31,32],challeng:[23,38],chang:[11,19,21,34,35,43],chaoyang:[21,25,33,41,43,46],chaoyangh:[21,41,43],chaoyanghe2020fedml:[41,46],charact:[33,36],chart:[10,11],check:[4,8,17,18,19,20,21,26,27,33,36,41,43,47],checkout:43,cheetah:[41,44,46],chmod:43,chosen:27,chrome:0,cifar100:[27,36],cifar10:[6,27,28,36],cifar10_dir:6,cifar:[6,27,33,36],cinic10:[27,36],citi:[16,18,20],clang:43,class_num:[4,10,11,31,32,35],classif:[31,32,35,38],clean:0,cleaningup:10,cleanup_pg:10,clearer:1,cli:[10,11,16,21,27],click:[6,19,21],client1:[10,11],client2:[10,11],client:[6,8,9,10,11,16,21,27,31,32,35,38,41],client_dist_launch:10,client_id_list:[10,11,28,32,34,41],client_index:32,client_login:21,client_manag:[10,11,31],client_master_manag:10,client_model_s3_address:11,client_num:[10,11,31,32,35],client_num_in_tot:[10,11,27,28,31,32,34,41],client_num_per_round:[10,11,27,28,32,34,41],client_numb:35,client_optim:[10,11,27,28,32,34,41],client_silo_config_path:10,client_slave_manag:10,clone:43,cloud:[16,18,20,21,38,41,43,47],cloud_region:[19,21],cluster:[2,23,33],cnn:[2,23,27,36],co:[16,23,41,46],code:[1,2,4,6,8,10,11,16,18,20,21,23,26,31,32,33,35,47],codedprivateml:23,codedreduc:23,collabor:[10,11,18,20,23,38],collect:[23,33,36,38],color:47,com:[1,4,6,8,11,17,18,20,21,22,26,31,32,35,36,40,41,42,43,46,47],come:[15,16,41,43],comm_arg:[10,11,31,32,41],comm_round:[10,11,27,28,32,34,41],command:[10,11,21,31,43],comment:[4,8,26],commerci:38,commit:43,common:[41,46],common_arg:[10,11,32,41],commun:[1,2,16,19,21,23,28,31,32,33,34,38,46],compani:[41,46],compar:[33,38],comparison:[1,2,38],compat:[4,43],competit:38,compil:[],complet:[1,21,31,32],complex:[10,11,18,19,20,21,41,43],compon:1,comprehens:[33,36,38],compromis:23,comput:[2,16,23,36,41,46,47],computation:38,concern:38,conda:[0,31,32,43],confer:[23,27],config:[4,10,11,17,21,31,41],config_fold:[17,21],config_vers:[19,21],configur:[1,21,31,32,41],congratul:[19,21],connect:[6,10],consid:1,consist:[1,6,16,33,36],consol:[6,21],constrain:38,constraint:2,construct:[1,9],contain:[10,33,36,38,43],contribut:[2,25,41],conv:[27,36],convent:2,convers:1,convinc:33,copi:[19,21,31],copyright:43,core:[2,16,23,38,41,46,47],correct:[23,43],correctli:[6,21],correl:23,correspond:[6,10,11,32],cost:[6,38],counterpart:1,coupl:19,cover:[41,44],cpp:42,cpu:[29,31,32,41],creat:[8,10,11,16,19,26,31,32,34,36,41,42,43],create_mnn_lenet5_model:4,create_model:32,creation:[19,21],credit:43,cross:[4,10,11,16,18,20,21,29,30,31,36,38,39,41,44,46],cross_devic:[4,6,41],cross_silo:[8,10,11,17,21,41],cryptographi:23,cs:6,cubic:23,cuda:43,current:[1,4,6,16,19,21,31,32,41,43],current_running_dir:[19,21],custom:[1,8,16,19,26,41,43,46],custum_data_and_model:[10,11],cut:[16,23,41,44,47],cute:33,cv:27,cvpr:23,cybersecur:[23,38],d7d0:10,d:[27,38],daemon:19,dalla:[18,20],darwin:43,data:[4,8,9,18,20,21,23,26,27,34,36,38,41,44,46],data_arg:[10,11,28,32,41],data_cache_dir:[10,11,31,32,35,41],data_load:[32,35,41],data_silo_select:10,datacent:43,dataload:[4,35],datapath:6,dataset:[2,6,8,10,11,16,26,27,28,29,31,32,34,38,39,41,43,46],dataset_nam:[10,11,31,32,35,41],ddp:[9,10,41],deal:31,debug:[6,21,43],decai:27,decentr:[1,16,18,20,23,33,34,38,41],decentril:39,decomposit:23,deep:[1,23],def:[1,10,11,31,32],defens:23,defici:23,defin:[1,10,31,32],definit:[4,8,26,31,32],degre:38,deliveri:1,demand:[8,26,40,41],demo:[16,18,20,33,41],demonstr:[1,10,33,38],depend:[31,32],deploi:[16,21,33,41],deploy:[16,18,19,20,33,41],deprec:35,describ:[1,19,21],descript:27,deserv:38,design:[1,2,8,16,23,26,31,38,41],desktop:42,despit:38,dest:[17,21],dest_fold:21,destin:21,destination_package_fold:17,detail:[1,2,4,8,10,11,18,20,26,27,32,33,34,41,47],detect:38,determin:[1,35],dev:[19,42],devel:43,develop:[1,2,4,6,8,16,26,29,30,33,38,41,43],devic:[1,4,8,10,11,16,19,21,26,31,32,34,38,39,41,43,44,46],device_arg:[10,11,32,41],device_id:[19,21],df:[17,21],dictionari:35,differ:[1,2,10,16,18,20,31,33,34,36,38,41],differenti:23,difficulti:[10,11],dimens:[32,35],direct:38,directli:[31,38,43],directori:[6,19,21,31,32],dirichlet:27,disconnect:10,discuss:[40,41],dispers:23,displai:16,dist:21,dist_training_arg:10,distribut:[1,2,9,10,11,16,19,21,25,27,38,41,44,46,47],divers:[1,2,16,23,41],dnn:[2,23],doc:[21,41,42,43],docker2:43,docker:[],document:[16,20,33],doe:[1,16],domain:[38,41],don:33,downgrad:38,download:[6,19,21,31,42],download_mnist:[10,11,31,32,41],dr:[33,41,46],due:[1,2,38],dure:[1,43],e1k05jd1tyw:41,e2b49004db48:10,e798061d62560b03e049d514e7cc8f1a753fde6b:43,e:[1,2,19,21,27,29,30,38,41,43],each:[1,4,6,8,9,10,11,19,21,26,31,32,33,34,35,36,41],earth:[18,20],easi:[18,20],easier:43,easili:2,echo:[31,43],econom:23,ecosystem:[16,37],edg:[6,16,18,19,20,21,23,38,41,44,46,47],edge_case_exampl:36,edge_deploy:21,edge_id:[11,19,21],edit:[32,43],edu:6,ee0b5f53d949c84cee7decbe7a619e63fb1f8408:10,ee0b5f53d949c84cee7decbe7a629e63fb2f8408:[11,32],efa:[29,30,41],effect:38,efficaci:38,effici:[16,18,20,23,38,41],effort:38,egg:41,either:[1,19],elast:23,email:[19,21],embed:[41,44],emnist:[27,33,36,39],empir:38,en:42,enabl:[2,16,18,19,20,23,34,38,41],enable_wandb:[4,10,11,32,41],encapsul:[10,11],end:[10,11,23,31,32,38],engin:[16,18,20,33,41],english:[33,36],enjoi:[19,21],entangl:23,entir:[1,32],entiti:4,entri:[17,21,41,43],entry_argu:[],entry_config:[],entry_point:21,entry_point_fil:17,env:[10,11,31,41,43],environ:[17,31,32,33,38,43],environment_arg:10,ep:[17,21],epoch:[10,11,27,28,31,32,34,41],error:31,errorcod:31,essenti:[2,41],etc:[9,16,19,23,29,30,32,38,41],evalu:[27,38],ex:27,exactli:31,exampl:[1,6,8,16,21,26,31,32,34,35,39,43,46],exce:[22,41],except:10,exchang:1,excit:38,exdb:6,execut:[1,10,31],exist:[1,19,21,33],exit:21,exp:28,expand:41,experi:[1,6,27,33,38,41,44,47],experiment:[16,19,27,28],explor:38,expos:38,expr:31,extend:[33,36],extens:38,facebook:[19,21],facilit:[38,41],fair:[2,23],fairf:23,fairli:33,fals:[4,10,11,32,41],faq:16,far:38,fast:23,fastest:[16,41,44,46],fc:27,feat:[25,41],featur:[4,8,19,21,26],fed_cifar100:[28,36],fed_emnist:36,fed_shakespear:[28,36],fedavg:[1,10,11,27,28,31,32,38,39,41],fedavg_api:32,fedavg_cross_silo:[],fedavgclientmanag:31,fedcv:[23,41],feddetect:38,feder:[1,5,10,11,16,18,20,21,25,29,31,32,33,39,41,43,44,46,47],federated_optim:[10,11,28,32,34,41],fedgkt:[33,34],fedgraphnn:[23,41],fediot:41,fedml3:21,fedml:[20,22,23,25,27,28,31,32,34,35,36,37,38,39,44,46,47],fedml_0_0_1:10,fedml_0_0_1_eecbc3ba:10,fedml_0_0_2:10,fedml_0_0_2_8792eda0:10,fedml_0_1_06180ac:11,fedml_0_2_565e48cb:10,fedml_0_2_a33a50ad:11,fedml_aggreg:[6,10,11],fedml_batch_bootstrap:43,fedml_batch_entry_script:43,fedml_client_manag:11,fedml_config:[10,11,31,32,34,41],fedml_cor:1,fedml_data:19,fedml_docker_imag:43,fedml_edge_deploy:19,fedml_experi:[],fedml_main_node_index:43,fedml_main_node_private_ipv4_address:43,fedml_node_index:43,fedml_num_nod:43,fedml_product:21,fedml_run_id:43,fedml_server_manag:10,fedml_startup:43,fedml_torch_fedavg_mnist_lr:[10,11,32],fedml_user_cod:17,fedna:[23,33,34],fednlp:[23,41],fednova:[33,34],fedopt:[33,34,38,39],fedsi:[],fedsys_client_id_list:[],fedsys_is_using_local_data:[],fedsys_private_local_data:[],fedsys_run_id:[],fedsys_synthetic_data_url:[],fedvg:[33,34],feedback:[18,19,20,21],feel:[40,41],femnist:28,few:[8,26,41],field:23,fig:1,figur:[1,9,18,19,20,21,27,33,34,41],file:[4,10,21,31,32,41],fileand:31,filenam:31,fill:[10,11,19,41],financ:16,financi:23,find:[6,21,23,43],finish:[1,10,21,31,32],finsihd:10,first:[1,10,11,31,32,41,42],five:[1,8,26,41],fl:[1,2,4,8,16,19,29,30,31,32,33,34,36,38,41,44,46,47],fl_client:21,fl_client_ag:[],fl_server:[],fl_server_ag:[],flclient_agent_266:21,flexibl:[1,2,8,19,26,31,41],flow:[1,10,11,21,25,31,41],flserver_ag:21,fntml:23,focu:[41,44],focus:1,folder:[6,10,11,19,21],follow:[1,6,8,10,11,16,18,19,20,21,22,26,27,31,32,33,34,35,36,37,39,41,42,43,46],forecast:[16,23],forens:23,forg:0,form:[4,8,26,35],formul:38,forward:[10,11,31,32,38],found:[19,31,41],foundat:[38,41],founder:[16,41,46],four:[18,20,33,36,38],fpcjrkh3bpy:[18,20],fraction:[10,11,31,32],framework:[1,4,10,11,23,26,32,33,34,35,39,41],free:[40,41],frequenc:38,frequency_of_the_test:[10,11,32,41],frequent:[40,41],fri:10,friendli:[1,16,41,44],from:[1,4,8,9,10,11,16,18,19,20,21,23,26,27,28,31,33,34,36,38,41],full:[23,41],fundament:[16,23],funni:[18,20],further:[10,38,47],furthermor:[10,16,38],futur:38,g:[1,2,19,21,29,30,38,41],gap:[2,33,38],gener:[1,31,41,46],generate_topolog:1,geo:19,get:[1,10,16,19,21,32,33,43,46],get_devic:[4,8,10,11,26,31,32,34,41],get_in_neighbor_idx_list:1,get_in_neighbor_weight:1,get_out_neighbor_idx_list:1,get_out_neighbor_weight:1,git:[42,43],github:[1,4,8,11,17,21,26,31,32,35,36,40,41,42,43,46,47],give:[1,31],given:38,global:[9,35,38,41],global_aggregated_model_s3_address:6,global_model:[10,11,41],global_model_file_path:[4,10,11,41],gnn:38,go:6,goal:10,good:33,gpu:[2,9,10,16,29,30,31,33,41,43],gpu_id:[32,41],gpu_map:[10,11,41],gpu_mapping_fil:[10,11,41],gpu_mapping_kei:[10,11,41],grade:[41,44],gradient:23,gradiveq:23,graph:[23,41],green:[6,47],grep:43,group:[10,11,16,18,20,23,27,36,41],grow:38,grpc:[39,41,46],guarante:23,guid:[6,10,11,16,23,42,43],guidanc:41,gz:6,ha:[1,8,10,26,33,38,41,42,46],handl:[1,2,18,19,20,21,33,41],handle_message_receive_model_from_cli:10,handle_message_receive_model_from_serv:10,handshak:[],hang:[41,46],happi:[40,41],have:[1,6,10,16,19,21,31,32,33,41,43],haven:6,havwx1e:[22,41],he:[25,33,41,46],healthcar:16,heck:34,help:[2,16,21,29,32,33,38,41,43],henc:[16,18,20,41],here:[9,10,11,17,31,32,41,43],herein:38,herg:38,hesit:33,hetero:[10,11,28,32,41],heterogen:[23,27],heterosag:23,hierarch:[1,10,16,23,33,34,41],hierarchi:10,hierarchical_fedavg_mnist_lr_exampl:[9,41],high:[16,29,30,38,41,47],highli:[10,11],highlight:1,him:[22,41],hire:[16,41],home:43,hongyi:[41,46],hope:[2,29,32,41],horizont:[16,41],hospit:[9,41],host:[31,33,36,43],hostfil:31,hostnam:31,how:[6,9,10,11,16,18,19,20,31,32,39,41,42],howev:38,html:[0,21,41,42,43],http:[1,4,6,8,11,16,17,18,19,20,21,22,26,27,28,31,32,35,36,38,40,41,42,43,46,47],hub:43,huge:38,human:33,hyper:[19,27,41],hyperparamet:27,hyperparameters_config:[],i:[31,38,43],iclr:[23,33,34,38],icml:23,id:[1,6,10,11,19,21,27,31,41,43],idea:[29,32,33,41,46],idx1:6,idx3:6,idx:6,ieee:23,ifconfig:10,ignor:1,iid:[27,38],ijcnn:23,ijcv:23,illustr:[1,9,10,11,21,41],imag:[6,10,11,23,31,32,33,36,38,43],imagenet:36,immedi:43,impact:[16,41],implement:[1,2,16,23,27,31,32,41,46],impli:38,improv:38,in_featur:32,in_neighbor_idx_list:1,in_neighbor_weight:1,inc:[16,33,41,43],includ:[1,10,11,21,31,32,33,34,36,38,41,44],increas:38,index:[0,1,10,11,31,32,35],indic:[23,32,38],industri:[16,41,44],infer:[23,41,46],infiniband:[29,30,41],inflex:1,info:[6,10,11,31,32,35,41],inform:[1,23,31,32,41,43,46],infrastructur:33,inher:38,inherit:1,init:[4,8,10,11,26,31,32,34,41],initi:[1,31,32],innov:[1,23],input:[19,21],input_dim:[10,11,31,32],insid:[1,9,10,41],instal:[10,11,16,31,32],instead:31,instruct:[6,19,21,32,33,42,43],insur:16,integ:31,integr:41,interact:43,interest:[16,38,41],interfac:[1,2,10,16,38,41],intern:[16,23,41],internet:[23,41],interpret:43,intrigu:38,introduc:[1,37,38],introduct:[16,41],invit:[10,11,16,22],invok:31,io:[41,42,44],iot:[23,38,41,43,44,46,47],ip:10,ipdp:23,is_mobil:31,isit:23,isregist:19,issu:[42,43],iter:43,its:[1,10,16,38],itself:[1,4,8,26],japan:[18,20],java:43,jetbrain:43,jinhyun:[41,46],join:16,journal:[23,41,46],jsait:23,just:[31,32,33,41],k0nze:42,k:[10,11],kaggl:[33,36],kang:[41,46],kdd:[23,38],keep:[4,8,26,38,41],kei:[10,23,35],kill:[31,43],know:16,knowledg:23,kriz:6,l:31,label:[1,6,23],lack:38,lagrang:23,land:23,landmark:36,languag:[16,23,36,41],laptop:[29,32,33,41],larg:[2,33,38,41,47],last:28,latenc:23,latent:27,later:[19,35],latest:[4,8,26,27,42,43],launch:[19,43],launcher_rdzv_port:10,law:27,layer:[23,35,36],lda:27,leakag:23,learn:[1,2,5,10,11,18,20,21,23,25,29,31,32,33,34,39,41,43,44,46],learning_r:[10,11,28,32,34,41],lecun:6,led:2,left:47,len:6,lending_club_loan:36,length:10,let:[19,21,31,41],level:[1,23,41,46],li:[41,46],lib:41,librari:[1,21,23,25,31,38,42,46,47],licens:43,life:43,lightsecagg:23,lightweight:[16,18,20,23,41,46],like:[2,19,21,41,43,46],limit:23,line:[1,6,8,26,41],linear:[10,11,31,32,36],linearli:23,link:[19,21,27,43],linux:[6,31,41,43,44],list:[1,6,10,19,21,27,33,36,41,43,47],literatur:38,liu:[41,46],live:[20,33,41,46],lkfg3lb13u5:27,ll:37,lo:[18,20],load:[4,8,10,11,26,31,32,34,35,36,41],load_cross_silo:10,load_data:[10,11,31,32,35,41],load_for_simul:32,load_partition_data_mnist:[10,11,31,32],load_synthetic_data:41,loader:[10,11,16,19,31,32],local:[9,16,19,35,38,41],local_log_output_path:10,local_serv:21,local_test_on_all_cli:32,locat:[6,10,11,21,35,43],log:[10,11,19,21,31,32,41],log_file_dir:[10,11,19,21,32,41],logger:[10,11,31,32,35],login:[10,11,16],logist:[27,29,32,33,36,41,43],logisticregress:[10,11,31,32],logo:41,logout:16,look:[31,43],loss:[6,10,11,31,32],low:[1,23,41,46],lower:[33,36],lr:[10,11,27,28,32,41],lstm:[27,36],m:32,machin:[1,2,16,18,20,23,32,41,44,46],maco:[6,19,21,43],mai:[1,10,19,21,31,33,38,41,43],main:[31,32,41,43],main_fedavg_cross_silo:[],maintain:[16,38,43],make:[0,1,32,38,43],manag:[10,11,16,41,46],mani:27,mapping_default:[10,11,41],mapping_serv:10,mapping_silo_1:10,mapping_silo_2:10,mar:43,march:[41,46],market:23,massiv:38,master:[1,4,8,10,11,17,21,26,31,32,35,36,41,43,46],master_address:10,match:10,matrix:23,maxdepth:39,maximum:[16,18,20,41],mb2:19,mbp:[21,43],md:[39,43],mean:[6,9,19,31,32,41,43],meaning:38,mechan:1,meet:[1,8,26,41],member:[19,21,22,41],memlock:43,memori:[2,23,38],merkl:23,messag:[10,11,16,21,41],message_json:[6,11],message_kei:[10,11],messeng:[19,21],mest:23,met:[42,43],method:[1,23,27,31,33,41],metric:27,mi:[41,46],microsoft:42,migrat:21,milena:23,mimick:33,mine:[16,23],minist:[31,32],minut:41,mission:[16,41],mitig:[23,33,38],mix:23,mkdir:6,ml:[23,25,27,33,41],mlop:[33,44,46],mlops_metr:[6,11],mlsp:23,mlsy:[23,25,41],mnist:[6,27,29,31,32,33,35,36,39,41,43],mnist_dir:6,mnistdataset:4,mnn37:[21,41,43],mnn:4,mobil:[4,47],mobilenet:[27,36],mode:[6,31,43],model:[2,4,8,18,19,20,21,23,26,28,29,34,38,41,43,44,46],model_arg:[10,11,28,32,41],model_config:[],model_dict:6,model_file_cach:[10,11,41],model_file_cache_fold:[10,11,41],model_hub:32,model_nam:32,model_train:32,modelnam:[],modern:2,modifi:[11,19,31],modul:[10,11,31,32,41,46],monitor:[10,11,19,21],more:[1,2,10,11,16,19,20,21,32,33,34,35,38,43],moreov:38,most:[1,38],move:6,mpi4pi:[1,31,42],mpi:[1,16,33,39,41,44,46],mpi_abort:31,mpi_base_framework_exampl:31,mpi_comm_world:31,mpi_decentralized_fl_exampl:31,mpi_host_fil:31,mpi_torch_fedavg_mnist_lr_exampl:[29,31,41,43],mpirun:31,mqtt:[10,11,39,41,46],mqtt_config:[10,11,41],mqtt_config_path:[10,11,41],mqtt_s3:[10,11,41],mqtt_s3_comm_manag:6,mqtt_s3_fedavg_hierarchical_mnist_lr_exampl:10,mqtt_s3_fedavg_mnist_lr_exampl:[6,9,11,21,41],mqtt_s3_multi_clients_comm_manag:[10,11],mqtt_s3_status_manag:10,msclkid:42,msg:[10,11],msmpi:42,multi:23,multipl:[9,10,23,38,41,43],multiprocess:1,murali:[41,46],mv:6,my:[6,19,21],my_dataset:4,my_model_trainer_classif:[10,11,31,32],mymodeltrain:32,myst:0,mysteri:38,n:[1,32],n_dist_train:10,n_node_in_silo:10,n_proc_in_silo:10,n_proc_per_nod:10,na:23,naacl:[23,38],name:[4,19,21,38,39,42,43],namespac:[19,21],nativ:43,natur:[16,23,36,41],nccl:[16,41,44,46],necessari:[1,38],necessit:38,need:[1,2,10,11,16,18,19,20,21,31,32,33,38,41,43],neighbor:1,neighbor_num:1,neighborhood:1,nerdschalk:42,net:[23,27,43],network:[1,23,41],network_interfac:10,neural:[2,23,41,46],neurip:[23,25,33,34,41,46],newli:[4,8,19,21,26,33],newton:23,next:[6,21,33,36,43],nlp:[27,38],nn:[1,10,11,31,32],node:[1,10,16,18,20,41],node_address:10,node_index:1,nofil:43,non:[2,27,31,33,38],none:[4,6,10,21,31],nonetyp:10,normal:[1,27,35,36,43],note:[1,10,11,17,19,21,27,28,31,43],notifi:10,novel:2,now:[19,20,21,31,32,41,43,46],np:31,number:[1,6,8,10,11,19,26,27,31,35,38,41],nus_wid:36,nvidia:43,nvlink:[29,30,41],nwp:36,object:[10,31,32,38],obtain:38,occur:38,octopu:44,offici:31,omnilyt:23,on_disconnect:10,on_messag:10,onc:[21,32],one:[8,26,31,32,41],one_lin:[10,11,21],ones:[19,38],onli:[1,2,4,8,21,26,31,41],open:[0,6,16,18,19,20,21,23,31,33,38,44,47],openreview:27,oper:[4,6,31,32,41,42,44,46],opportun:23,opt:41,optim:[2,23,27,38,39],option:[6,21],orchestr:16,order:[1,10],org:[1,27,38],organ:[16,38,40,41,44,46],orient:[2,33,36],origin:[10,11,31,32,33,36,41,43,46],os:43,os_nam:[19,21],other:[1,19,21,31,32,33,41,43,44,46],our:[1,2,4,8,10,11,16,18,19,20,21,26,27,31,33,38,41,42],out:[1,2],out_directed_neighbor:1,out_featur:32,out_neighbor_idx_list:1,out_neighbor_weight:1,output:[6,10,11,21,31,32,35,41],output_dim:[8,10,11,26,31,32,34,41],over:[1,16,23,31,38],overdesign:31,overflow:[33,36],overhead:38,overview:[],overview_training_ori:1,own:[1,16,19,21,43],p:6,pachinko:27,pack:10,packag:[16,19,31,32,41,43,46],page28:27,page34:27,page:[6,19,21,27],paper:[10,11,16,25,27,28,31,32,36,38,41,46,47],paradigm:[2,38],paradistributeddataparallel:1,paragraph:27,parallel:[1,10],param:[10,35],paramet:[1,19,27,38,41],parrot:[8,34,35,44,46],pars:19,parser:0,part:[1,31,32],partial:23,particip:1,particularli:38,partit:[19,27,38],partition_alpha:[10,11,28,32,41],partition_method:[10,11,28,32,41],pass:[1,16,21,41],password:[18,20],passwrodless:10,path:[1,10,19,21],patit:27,patition_alpha:27,pattern:[1,2,31],payload:31,pdf:[27,38],pdsh:10,peilin:[41,46],peopl:16,per:38,perform:[10,11,19,21,23,33,36,38,39],permiss:43,person:[18,20,23,29,32,38,41],perspect:1,pet:33,pg:10,phase:43,philosophi:[1,8,26,31,41],pi:[38,43],pick:19,pip:[19,31,32,33,41,42],pipe:23,pipelin:[23,41,44],pipetransform:23,plan:11,platform:[16,18,19,20,33,38,41,46],pleas:[4,6,8,10,11,16,18,19,20,21,22,26,27,31,32,33,34,35,36,40,41,42,43,47],plugin:33,point:[2,21,38,41],polynomi:23,polyshard:23,popular:[1,38],port:10,possibl:[8,26,38,41],post:[42,43],pow:27,power:27,practic:2,praneeth:[41,46],prcoess:10,pre:[],predefin:1,predict:[23,33,36],prefer:19,prepar:[6,20,43],preprocess:38,present:[10,31,32,38,41,46],preserv:[23,38],press:21,print:1,privaci:[1,16,18,19,20,38,41,46],privat:[18,20,23],private_config_path:10,privileg:43,problem:23,proce:31,procedur:1,process:[1,10,16,19,21,23,31,33,34,35,36,39,41,44,46],process_num:31,processor:42,product:[16,17,38,41],professor:33,profil:[10,11],program:[2,31,32,43],progress:[19,21,38],prohibit:38,project:[1,4,10,11,16,41,42,43],promis:38,prompt:43,propos:38,protect:23,protocol:1,provabl:[16,18,20,41],provid:[1,10,11,16,18,19,20,21,32,33,38,41,44,46],ps:43,pt:[10,11,41],publicli:38,publish:[33,38],pull:43,purpos:[41,46],push:[6,38],py3:41,py:[4,6,8,10,11,17,21,26,31,32,35,41,43],pycharm:43,pypi:1,python3:[10,11,41],python:[4,6,8,10,11,16,21,26,31,32,35,36,41,42,43,44,46],pytorch:[1,9,10,35,39,41,46],q:43,qiang:[41,46],qiu:[41,46],qr:6,quadrat:23,quantiz:23,queri:1,question:[33,36,38,40,41,42],quick:[29,32,33],quickli:[31,32],r34w:[22,41],r:[27,31],ramesh:[41,46],random_se:[10,11,32,41],rang:[16,38],rank:[10,11,31,41],rapidli:38,rare:38,raskar:[41,46],raspberri:[38,43],rate:[23,27,38],rather:1,read:[10,11,31,32,33,35,41],readi:[10,19,21],readm:43,real:[10,11,21,23,27,31,33,38,41,44],realist:[2,27,38],realli:36,receiv:[1,9,10,41],receive_id:10,recent:38,recommend:[10,11],reduc:[8,10,11,26,41],ref:[1,28],refactor:[6,35,36,41,43,46],refer:[1,6,10,11,16,27,28,32,38,43],referenc:28,reference_exampl:1,reflect:[1,41,43],reformul:23,refresh:21,regist:21,register_message_receive_handl:1,registr:21,regress:[27,29,32,33,36,41,43],regul:38,rel:19,relat:[1,4,16,19,21,27,31,32,33,41,46],releas:[6,17,19,21],release_401:43,relev:6,rememb:[6,43],remot:[1,43],remov:35,rendezv:10,reorgan:[41,46],replac:21,report:[27,31],report_aggregated_model_info:6,report_client_model_info:11,report_server_training_metr:[10,11],report_server_training_round_info:11,repres:[23,31,33,34,36,38,41,47],represent:38,reproduc:27,requir:[1,6,43],research:[1,2,16,18,20,29,31,32,33,36,38,41,44,46,47],resili:23,resnet18:[27,36],resnet18_gn:28,resnet56:28,resnet:[27,29,32,36,41],resolut:23,resourc:[4,8,26,33,38,46],respect:10,respons:1,rest:[],restart:43,restrict:[1,38],result:[1,9,19,21,33,38,41],reus:[2,47],review:[19,21,23],rf:6,rich:33,right:6,ring:1,rm:[6,43],rnn:[27,36],robust:[23,39],root:32,round:[6,10,11,23,27,28,31,32,38],round_id:[10,11,31],round_idx:[6,11],round_index:11,round_numb:10,rpc:[1,41,46],rtx2080tix4:43,run:[0,4,6,8,9,10,11,16,19,26,28,31,32,33,34,41,44],run_client:[10,11],run_cross_silo_cli:11,run_cross_silo_serv:11,run_hierarchical_cross_silo_cli:10,run_hierarchical_cross_silo_serv:10,run_id:[6,11,41],run_nam:32,run_one_line_exampl:[31,43],run_serv:[6,10,11],run_simul:[8,26,31,32,41],run_step_by_step_exampl:31,running_tim:11,s3:[10,11,39],s3_config:[10,11,41],s3_config_path:[10,11,41],s:[10,11,19,21,23,31,32,35,41,44,47],salman:[25,41,46],same:[1,10,31,32,43],sampl:[10,11,31,32,35],saniti:[17,33],save:[],scalabl:[23,25,41],scale:[2,16,18,20,33,38,41,44,46,47],scenario:[2,10,16,41,46],schedul:[23,38],scientif:[16,41],scm:42,scope:16,screenshot:[19,21],script:[6,10,11,19,21,43],sdcard:6,sdk:[4,16,41,44,47],seamlessli:[16,21],search:23,sec:1,second:1,section:[1,16,27,31,32],secur:[1,16,18,20,38,41,46],sed:31,see:[1,6,10,11,18,19,20,21,31,32,37,43],segment:38,select:[19,21,23],self:[1,6,10,11,23,31,32],semi:23,seminar:[25,41],send:[1,10,11,19,21,31],send_messag:[1,10,11],send_message_sync_model_to_cli:10,sens:41,sensi:[23,38],sensor:38,sent:[10,11],separ:[1,2],seq2seq:38,sequenc:[23,38],seri:[16,23,38],serial:[6,23],serv:[10,11,21,23,33,41,44,46],server:[4,8,9,10,11,16,21,29,31,32,41,43],server_config_path:10,server_gpu_mapping_kei:10,server_manag:10,serverless:23,servermnn:4,servic:[],set:[1,10,19,27,31,32,38,41,43],setup:43,sf:[17,21],sgd:[1,10,11,23,27,28,32,34,41],sh:[0,6,10,11,17,31,43],shakespear:[27,29,32,33,36,41],shallow:[10,11,31,32],shard:23,share:[38,41,46],shared_invit:[22,41],shen:[41,46],shm:43,should:[1,6,10,19,21,31,32,43],show:[1,6,18,19,20,21,39,41],showcas:38,shown:[1,9,10,11,19,21,31,32,33,34],shuffl:4,side:[10,11,47],sigmoid:[10,11,31,32],sign:21,signal:23,signific:38,silo:[9,10,11,18,20,21,36,39,41,44,46],silo_1:10,silo_2:10,similar:[1,6,31,32],simpl:[21,34,39,41,44],simplest:[1,35],simplic:[8,26,41],simplifi:[16,18,20,33],simul:[2,26,33,34,39,41,43,44,46],simulation_sp:41,simulatormpi:[26,31,34],simulatorsingleprocess:34,simultan:[8,23,26,41],singh:[41,46],singl:[10,16,31,33,34,39,41,44,46],single_process:41,site:41,size:[10,11,27,31,32,43],skeleton:33,slack:[16,19,21,41],slave:10,small:[23,29,32,41],smart:16,smartphon:[5,41,44,46],smoothli:[33,41],snippet:1,so:[32,41,46],social:[18,20],sock:43,softwar:[31,47],solut:[16,38],solv:[23,38],some:[1,38,41,43,46],someth:43,songz:[41,46],soon:[6,15,19,21,43],sourc:[1,4,6,8,10,11,16,17,21,26,35,44,47],source_fold:[17,21],sourcecod:[21,43],sp:32,sp_fedavg_mnist_lr_exampl:[29,32,41],spars:23,sparsiti:38,speak:33,special:[40,41],specif:[1,2,6,10,16,19,21,31,32,38,41,43],specifi:[6,10,21,31,32,34,43],sphinx:0,spider:23,split:[1,10,19,33,34,38],split_data_for_dist_train:10,spreadgnn:23,ssfl:23,ssh:10,stack:[23,33,36,41],stackoverflow:[27,36,42],stackoverflow_lr:28,stackoverflow_nwp:28,stage:16,stand:[16,27,31,32],standalon:[1,2,16,31,41],standard:1,stanford:[25,41],start:[4,8,10,11,16,19,26,31,32,33,34,46],start_run:[],start_train:21,starter:41,statist:2,statu:[10,11,19,21],step1:[31,32],step2:[31,32],step:[6,19,21],still:[31,32],stop:10,stop_receive_messag:10,stop_train:21,storag:38,str:1,straggler:23,strategi:38,structur:[35,38],student:33,studi:38,studio:43,submit:2,subscrib:21,subsequ:31,successfulli:[6,19,21,31,32],sudo:43,suggest:[2,29,30,38,41],suitabl:38,summar:27,summari:43,sun:41,supervis:23,support:[1,4,16,19,21,33,34,36,38,40,41,43,44,46],sure:32,swap:19,symmetr:1,symmetrictopologymanag:1,sync_process_group:10,synchron:1,synthet:[27,29,32,33,36,41],system:[1,2,10,11,16,19,21,33,41,42,46],systemat:[38,43],systemctl:43,t:[6,17,21,22,33,41,43],table1:27,table2:27,table7:27,table8:27,tacc:23,tackl:23,tag:[33,36,38,43],tail:36,tailor:[36,41,46],take:[10,35],talk:25,tangibl:33,tar:6,target:23,task:[23,27,32,33,35,36,43],teach:33,team:33,technolog:[16,23,41],tediou:[33,43],templat:[41,46],tencent:38,tensorflow:1,termin:[0,31],test:[6,10,11,17,28,31,32,35,42],test_acc:[11,32],test_data_glob:[10,11,31,32,35],test_data_local_dict:[10,11,31,32,35],test_data_num:[10,11,31,32,35],test_dataload:4,test_dataset:4,test_loss:[11,32],test_on_server_for_all_cli:[6,10,11],test_path:[10,11,31,32,35],text:[21,23,38],textit:1,texttt:1,than:[1,16,38],thank:[16,38,41],thei:43,them:[10,31],theoret:1,theori:23,therefor:10,thi:[1,2,6,10,11,16,17,19,21,29,30,31,32,33,36,38,39,41,42,43,46,47],thing:[23,41],though:[31,32],three:38,through:[1,6,10,11,21,31,32,35],thu:[1,6],tier:[27,33,36],time:[2,6,16,23,31,38],timestamp:11,titl:[41,46],tnnl:23,toctre:39,tokyo:[18,20],tool:[19,21,42],toolkit:43,top:[6,27,33,36],topic:10,topolog:[19,21,23,33,34,41,46],topologymanag:1,torch:[1,10,11,31,32],torch_client:[10,11,17,21],torch_fedavg_mnist_lr_custum_data_and_model_exampl:[31,32],torch_fedavg_mnist_lr_one_line_exampl:[31,32],torch_fedavg_mnist_lr_step_by_step_exampl:[31,32],torch_serv:[10,11,17,21],toronto:6,total:[10,31,35],total_round:11,tour:41,toward:23,tpmgr:1,track:[10,11,16],tracking_arg:[10,11,32,41],tradeoff:23,train:[1,2,4,6,8,9,18,19,20,26,34,35,38,41,44,46,47],train_arg:[10,11,28,32,34,41],train_data_glob:[10,11,31,32,35],train_data_local_dict:[10,11,31,32,35],train_data_local_num_dict:[10,11,31,32,35],train_data_num:[10,11,31,32,35],train_dataload:4,train_dataset:4,train_path:[10,11,31,32,35],trainer:[10,41,46],trainer_dist_adapt:10,training_acc:[10,11,32],training_loss:[10,11,32],training_num:6,training_typ:[10,11,32,41],transact:23,transfer:23,transform:[2,23,38],transplant:[16,33],tree:[1,11,21,23,31,32,35,36,41,46],trend:38,trick:38,trigger:[],trivial:[2,33],trpc:39,trustworthi:[25,41],tue:31,tune:41,turbo:23,turn:6,tutori:[16,19,39,41,42],two:[1,6,10,31,32,33,36,43],type:[1,10,11,19,21,31,38,43],u:43,ubuntu20:43,ubyt:6,uci:36,ui:[18,19,20,21,41],ulimit:43,under:[9,19,20,21,23,31,32,35,38],underli:1,understand:[2,38],undirect:1,undirected_neighbor_num:1,unifi:38,uniformli:[10,11,31,32],unique_device_id:[19,21],univers:38,unravel:38,unstructur:23,up:21,updat:[10,11,31,32,43],upgrad:[32,33,41,43,46],upload:[16,19,31],upper:[33,36],url:[19,21],us:[1,2,4,6,10,11,16,19,21,23,29,30,31,32,33,34,35,36,38,39,41,42,43,44,46],usa:[18,20],usag:[21,36,43],usb:6,user:[1,6,16,18,20,31,38,42,44,46],userid:[17,19],usernam:[18,20],using_gpu:[10,11,32,41],using_mlop:[10,41],usr:[10,11,31],util:[1,19,21,38,41,46],v:[17,18,20,41,43,46],valid:28,validation_arg:[10,11,32,41],valu:[19,21,32,33,35],valuabl:[2,19,21],variabl:43,variou:[1,16,18,20,38,41],vector:23,vepakomma:[41,46],verifi:23,versatil:[23,41,44],version:[1,2,16,19,21,31,32,41,43,46],vertic:[1,16,33,34],vfl:36,via:[16,19,23,33,36],video:[16,18,20],vision:[16,36,41,47],visionari:[23,41],visit:[18,20,27,33],visual:[10,11,19,21,42,43],visualstudio:[42,43],vldb:23,vmlldzoxode2ntu:27,wa:31,wai:31,wait:10,wandb:[4,27,28],wandb_ent:32,wandb_kei:[10,11,32],wandb_nam:[10,11],wandb_obj:4,wandb_project:[10,11,32],wang:[41,46],want:43,watch:[18,20,41,46],wd:27,we:[1,2,4,6,8,10,11,19,21,23,26,27,29,30,31,32,37,38,40,41,43,44],web:[6,18,19,20,21,41],websit:[33,36],wechat:[16,41],wed:11,weight:1,weight_decai:[10,11,32,34,41],welcom:[18,20],well:[6,33,47],wget:6,what:31,when:[1,21,29,31,32,35,41,43],where:[10,43],whether:43,which:[1,2,4,9,10,21,31,32,33,41,46],white:[27,41,46],whole:31,wide:[1,38],wider:38,william:[33,36],win:42,window:[6,10,11,31],wireless:23,wise:23,without:[1,2,16,18,19,20,21,33,35,41,43],won:[41,46],word:[33,36,41,44],work:[1,6,16,19,33,36,38],workabl:31,worker:[2,10],worker_num:[11,31,41],workermanag:1,workflow:[18,19,20,21,41],workshop:[23,25,38,41,46],workspac:43,world:[10,11,18,20,21,23,25,33,38,41,44],wors:38,would:[19,21,43],wrapper:41,write:[21,31],www:[6,18,20,41,43,46],x64:42,x86_64:43,x:[10,11,31,32],xda:6,xiaoyang:[41,46],xxx:[6,32],xzvf:6,y:43,yaml:[10,11,31,32,34,41],yaml_config_fil:41,yaml_path:41,yan:[41,46],yang:[41,46],yann:6,ybv29kak:28,ye:36,year:[16,23,33,41,46,47],yellow:1,you:[6,10,11,16,17,19,21,22,31,32,33,36,37,39,41,42,43],your:[16,19,21,29,30,32,40,41,43],youtub:[18,20,41,46],zero:[16,18,20,41],zhang:[41,46],zhao:[41,46],zip:[21,32,41],zoo:[41,46],zt:[22,41]},titles:["Install","FedML APIs (core)","FedML APIs (high-level)","<no title>","FedML BeeHive API Reference","FedML Beehive Examples","Federated Learning on Android Smartphones","FedML Beehive User Guide","FedML Octopus API Reference","FedML Octopus Examples","FedML Octopus Example with MNIST + Logistic Regression","FedML Octopus Example with MNIST + Logistic Regression","FedML Octopus User Guide","FedML Cheetah API Reference","Examples","FedML Cheetah User Guide","Welcome to FedML","FedML MLOps CLI and API Reference","FedML MLOps Live Tutorial","FedML MLOps User Guide","Tutorials for MLOps","FedML MLOps User Guide","Join the Community","Research Publications","Slides","Video","FedML Parrot API Reference","Benchmarking Results for MPI-based federated learning","Benchmarking results when using single process-based simulation","FedML Parrot Examples","Simulation with NCCL-based MPI (the fastest training)","Simulation with Message Passing Interface (MPI)","Simulation with a Single Process (Standalone)","FedML Parrot User Guide","Algorithm Reference Implementations","Customizing Data Loader for Your Own Dataset","Datasets and Models","Concepts","Ecosystem","<no title>","FedML FAQ","Get Started","FedML Installation on Windows OS","Installing FedML","Mission","FedML MLOps Video Tutorial","Open Source Code Architecture","Technology Overview"],titleterms:{"1":[6,17,19,21,31,32,47],"10":28,"2":[6,17,19,21,31,32,47],"3":[6,17,19,21,31,32,47],"4":[17,19,21,31,32],"5":21,"56":28,"new":31,"public":[23,41,47],A:[7,10,11,38,41,47],One:[10,11,31,32],The:[],account:19,account_id:21,adb:6,agent:21,ai:[10,11,17,23],algorithm:[23,31,33,34,47],allreduc:12,an:19,anaconda:43,android:[6,7,37,43],anywher:[],api:[1,2,4,8,10,11,13,17,26,31,32,36,41],apk:6,app:43,applic:[7,21,23,47],architectur:[41,46],autom:21,backend:12,base:[7,12,27,28,29,30,31],basic:37,beehiv:[4,5,7,16,41],benchmark:[27,28,33,38],better:[10,11],beyond:1,bind:6,build:[0,17,21],c:7,career:16,cheetah:[13,15,16],cifar100:28,cifar:28,cli:17,client:[7,17,19],cloud:[],cnn:28,code:[12,19,41,42,43,46],collabor:[16,19,21,47],command:6,commun:[12,22,41],comput:38,concept:37,conda:42,configur:19,connector:16,coordin:1,core:1,creat:21,cross:12,custom:[10,11,31,32,33,35],cv:36,data:[6,10,11,16,19,31,32,33,35],dataset:[33,35,36],decentr:31,definit:1,demo:7,deploi:[0,7],deploy:[21,47],design:47,develop:[21,31],devic:[6,12],differ:12,displai:17,distribut:[12,23],divers:[12,38],dnn:27,docker:[42,43],doe:21,download:36,ecosystem:[38,41],edg:[],edit:19,emnist:28,end:[],engin:7,exampl:[5,9,10,11,14,17,29,33,41],exampless:41,experi:[10,11,19],experiment:21,explan:[],faq:[40,41],fastest:[29,30],featur:41,fed:28,fedcv:38,feder:[6,9,12,23,27,28,36,37,38],fedgraphnn:38,fediot:38,fedml:[1,2,4,5,6,7,8,9,10,11,12,13,15,16,17,18,19,21,26,29,33,40,41,42,43,45],fedna:36,fednlp:38,financ:36,five:[10,11],fl:[6,7,9,12,23],framework:[31,38],friendli:47,from:[17,42,43],full:47,get:41,googl:28,gradient:1,graph:38,group:[19,21,22],guid:[7,12,15,19,21,33,41],guidanc:43,hierarch:[9,12],high:[2,23],horizont:[9,36],how:21,hyper:28,iid:28,impact:[23,47],implement:[33,34],industri:47,instal:[0,6,21,33,41,42,43],interfac:[29,31],internet:38,introduct:12,invit:[19,21,41],iot:16,join:[22,41],land:16,languag:38,laptop:6,larg:[16,23],leaf:28,learn:[6,9,12,16,27,36,37,38],level:[2,47],librari:[37,41],lightweight:27,line:[10,11,31,32],linear:27,live:18,loader:[33,35],local:[12,21],login:[17,19,21],logist:[10,11,28],logout:17,manag:[1,19,21],messag:[1,29,31],method:38,migrat:[],mission:44,ml:47,mlop:[6,7,10,11,16,17,18,19,20,21,37,41,45,47],mnist:[10,11,28],mobil:6,model:[1,10,11,16,27,31,32,33,36],modern:27,more:41,mpi:[27,29,30,31,42],natur:38,nccl:[29,30],network:[27,38],neural:[27,38],nlp:36,non:28,octopu:[8,9,10,11,12,16,41],onc:[],open:[10,11,17,37,41,46],optim:47,option:19,orchestr:7,orient:1,os:42,other:36,our:22,outlin:[16,23,33],overview:[16,17,41,47],own:[33,35],packag:[17,21],paper:23,paramet:[28,31,32],parrot:[12,16,26,29,33,41],pass:[29,31],path:6,pip:43,platform:[6,17,21,47],prepar:[31,32],preview:0,privaci:[23,47],privat:19,process:[28,29,32,38],program:1,project:[19,21],provid:36,python:7,quick:41,real:[12,16,47],realiti:16,recommend:43,recruit:41,refer:[4,8,13,17,26,33,34,41,46],regist:19,regress:[10,11,28],repositori:43,reproduc:28,research:23,resnet:28,resourc:[16,41],result:[10,11,27,28,31,32],rnn:28,run:[21,43],scale:[21,23],scenario:12,scientif:[23,47],sdk:[7,43],seamlessli:12,secur:[23,47],server:[6,7,17,19],servic:36,set:6,setup:[31,32],shakespear:28,shallow:27,silo:[12,16],simpl:16,simplifi:[21,41,47],simul:[12,16,21,28,29,30,31,32],singl:[28,29,32],slack:22,slide:24,smartphon:[6,16],sourc:[19,37,41,42,43,46],speedi:16,split:36,stack:47,stackoverflow:28,standalon:[29,32],start:[6,21,41],step:[31,32],succeed:43,support:12,synthet:[6,19],system:[23,38,47],talk:41,task:38,technolog:47,test:43,thing:38,tool:6,topolog:1,track:[19,21],train:[7,10,11,12,16,21,23,29,30,31,32],trainer:1,transfer:6,transplant:12,troubleshoot:43,tutori:[18,20,45],updat:19,upload:21,us:[27,28],user:[7,10,11,12,15,19,21,33,41,43,47],version:17,vertic:[9,36],via:21,video:[25,41,45,46],view:[31,32],vision:[23,38],visionari:47,wechat:22,welcom:16,when:28,why:16,window:[42,43],work:21,worker:1,workflow:[],world:[12,16,47],write:[],your:[6,33,35]}})
\ No newline at end of file
+Search.setIndex({docnames:["README","api_doc/api-core","api_doc/api-fedml-api","benchmark/index","cross-device/api","cross-device/examples","cross-device/examples/mqtt_s3_fedavg_mnist_lr_example","cross-device/user_guide","cross-silo/api","cross-silo/examples","cross-silo/examples/mqtt_s3_fedavg_hierarchical_mnist_lr_example","cross-silo/examples/mqtt_s3_fedavg_mnist_lr_example","cross-silo/user_guide","distributed/api","distributed/examples","distributed/user_guide","index","mlops/api","mlops/examples/mlops_live_demo","mlops/mlops_user_guide","mlops/t-mlops","mlops/user_guide","resource/community","resource/papers","resource/slides","resource/video","simulation/api","simulation/benchmark/BENCHMARK_MPI","simulation/benchmark/BENCHMARK_simulation","simulation/examples","simulation/examples/fedavg-mnist-examples","simulation/examples/mpi_torch_fedavg_mnist_lr_example","simulation/examples/sp_fedavg_mnist_lr_example","simulation/user_guide","simulation/user_guide/algorithm-reference-implementation","simulation/user_guide/data_loader_customization","simulation/user_guide/datasets-and-models","starter/concepts","starter/ecosystem","starter/examples","starter/faq","starter/get_started","starter/install/windows","starter/installation","starter/mission","starter/mlops_video","starter/oss_code_architecture","starter/overview"],envversion:{"sphinx.domains.c":2,"sphinx.domains.changeset":1,"sphinx.domains.citation":1,"sphinx.domains.cpp":4,"sphinx.domains.index":1,"sphinx.domains.javascript":2,"sphinx.domains.math":2,"sphinx.domains.python":3,"sphinx.domains.rst":2,"sphinx.domains.std":2,sphinx:56},filenames:["README.md","api_doc/api-core.md","api_doc/api-fedml-api.md","benchmark/index.md","cross-device/api.md","cross-device/examples.md","cross-device/examples/mqtt_s3_fedavg_mnist_lr_example.md","cross-device/user_guide.md","cross-silo/api.md","cross-silo/examples.md","cross-silo/examples/mqtt_s3_fedavg_hierarchical_mnist_lr_example.md","cross-silo/examples/mqtt_s3_fedavg_mnist_lr_example.md","cross-silo/user_guide.md","distributed/api.md","distributed/examples.md","distributed/user_guide.md","index.rst","mlops/api.md","mlops/examples/mlops_live_demo.md","mlops/mlops_user_guide.md","mlops/t-mlops.rst","mlops/user_guide.md","resource/community.md","resource/papers.md","resource/slides.md","resource/video.rst","simulation/api.md","simulation/benchmark/BENCHMARK_MPI.md","simulation/benchmark/BENCHMARK_simulation.md","simulation/examples.md","simulation/examples/fedavg-mnist-examples.md","simulation/examples/mpi_torch_fedavg_mnist_lr_example.md","simulation/examples/sp_fedavg_mnist_lr_example.md","simulation/user_guide.md","simulation/user_guide/algorithm-reference-implementation.md","simulation/user_guide/data_loader_customization.md","simulation/user_guide/datasets-and-models.md","starter/concepts.md","starter/ecosystem.md","starter/examples.md","starter/faq.md","starter/get_started.md","starter/install/windows.md","starter/installation.md","starter/mission.md","starter/mlops_video.rst","starter/oss_code_architecture.md","starter/overview.md"],objects:{},objnames:{},objtypes:{},terms:{"0":[1,6,10,11,21,27,28,31,32,34,41,43],"001":[10,11,27,32,34,41],"003":[27,32],"01":[27,41],"014680":32,"018868":11,"02":[6,31],"03":[11,27,28,32,34,41],"04":[6,10,11,31,32,43],"046":32,"051":32,"053":32,"053363":11,"054":31,"055":31,"057":32,"059":32,"060":32,"062":32,"064":32,"067":32,"068":32,"07":10,"070":32,"07145":38,"071570":32,"07976":38,"08815":38,"093762":32,"0x000001cf758d8f88":32,"0x9801a7a55e85":19,"0xb6ff42da6a7":21,"1":[1,10,11,16,23,27,28,34,39,41,43,44,46],"10":[4,6,10,11,23,27,31,32,34,41,42,43,47],"100":[10,11,27,31,32,33,36],"1000":[10,11,27,28,32,34,41],"104":31,"105":19,"107":11,"11":[23,42,43],"11066":38,"115":10,"116":10,"118":31,"12":[23,27,32],"1200":27,"124":[10,32],"127":[21,43],"128":[6,11],"128363870":32,"129":[6,10],"13":23,"130":32,"132":[10,32],"132013":32,"13518":27,"137":[10,11],"14":[23,32,41],"140":19,"14285715":1,"145":11,"15":[10,23,32],"1500":27,"155":10,"156245":32,"16":[10,23,27],"160":10,"164":[6,10],"165":10,"1651030718":11,"16666667":1,"168":10,"17":32,"172":10,"18":[27,28,29,32,41],"186":10,"189":6,"19":[27,28,32],"191":10,"1912692754254417":10,"192":10,"197":[10,32],"198":32,"199":[11,32],"1ba9":10,"1canbw":28,"1d4b":11,"2":[1,10,11,16,23,27,28,33,36,39,41,43,44,46],"20":[10,11,23,27,28,31,32],"200":[6,22,27,28,32,34,41],"2000":28,"2007":27,"201":6,"2018":23,"2019":23,"2020":[23,25,33,34,41,43,46],"2021":[23,33,34,38],"2022":[6,10,11,23,31,32,38,41,46],"204":10,"21":[23,27,32,38],"2104":38,"2106":38,"2111":38,"212":10,"212439":32,"215":10,"22":[6,23,31,41],"221":32,"222175":32,"224":10,"225":11,"23":27,"235":10,"237719":32,"240":11,"244":10,"246":10,"25":[10,27,32],"250":10,"253":10,"254":10,"257":10,"26":[1,31,32,43],"261":10,"262":11,"263015":10,"266":21,"269":10,"27":[6,11,27],"271503":10,"276":32,"276260":10,"28":[6,10,11,31,32,41],"284570":10,"284942":10,"288988":10,"29410":10,"295648":32,"296":[10,11],"296894":10,"29th":23,"2al5q5mi":28,"3":[1,10,11,16,23,27,28,33,41,42,43,44,46],"30":[10,27,31,32,33,41],"300":6,"301717":10,"302":21,"306":11,"306870":10,"308":[10,32],"31":[6,10,41],"317":32,"32":[6,27,43],"33":[10,31],"33333334":1,"337":10,"34":[28,32],"3400":27,"342477":27,"36":38,"369121":32,"37611896730669436":10,"3785d45dce8911ecbae4ea89f38f9f29":42,"38":[11,32],"3aponqml":28,"3lv4gmpz":28,"4":[1,16,23,27,43],"40":[10,11,27,32],"4000":[27,28],"41":32,"41b8":11,"42":32,"424e":10,"436":32,"44":[10,27,28,32],"444":32,"445c":11,"45":[11,32],"47":[10,11],"479":32,"48":[6,10],"485":32,"49":[6,10,11,27,31],"4ae765659471":11,"4b6d":10,"4btyrt0u":28,"4d9d":10,"5":[1,10,11,16,19,23,27,28,31,32,41,43],"50":[10,11,23,27,32,41,47],"500":27,"506":32,"50th":[10,11],"51":11,"516":32,"52":32,"53":[6,27,28,31,43],"54":27,"54615336":42,"55":[10,27],"56":[11,27,31],"57":[10,27,28,32,41],"5741":11,"58":11,"5836178064346313":6,"59":41,"591":11,"6":[1,19,23,27,43],"60":[27,32],"600":[33,36],"602":11,"62":32,"63":[11,31,32],"64":[4,27,42],"64g":43,"65535":43,"661":32,"67":[11,31],"68":27,"685":32,"6be8":10,"7":[1,21,23,27,28,31,38,41,42,43],"70":[27,32],"71":[27,31],"711":[6,32],"715":27,"7155714841722886":11,"7156":11,"717948717948718":11,"73":27,"7344120322737784":32,"73456031979222":32,"74":11,"75":[10,27,32],"750":32,"76":6,"77":6,"775815":31,"777":43,"784":32,"79":27,"794":32,"7pf2c9r2":28,"8":[1,23,27,28,41,43],"80":[10,28,32],"803107":11,"81":[11,28,32],"817":32,"818":32,"8188848188848189":32,"8190029839128179":32,"82":27,"83":6,"839799":31,"84":[27,28],"845405":31,"86":27,"860":32,"861":32,"866":32,"867":32,"867b":10,"87":27,"88":[6,32],"887":32,"888":32,"891":32,"8972983557921448":11,"8997":11,"8997359397010631":11,"8b7e880db4c5":10,"9":[23,27,28,31,43],"90":17,"905":32,"91":[6,27],"920599":31,"9205eea272f2":11,"925":32,"93":27,"930":11,"93setzgzmyi":46,"94":10,"942":32,"95":27,"954":32,"959164":32,"96":32,"966263":32,"973":32,"984373":11,"986":32,"9c71":10,"\u03b1":27,"\u03b2":27,"abstract":1,"break":23,"case":[29,30,31,32,33,36,41,43,44],"class":[1,10,11,31,32,33,35,36],"default":[19,41,43,46],"do":[1,10,11,18,19,20,21,31,33],"final":43,"function":[1,19,31,32],"import":[1,4,8,10,11,23,26,31,32,34,41,43],"int":1,"new":[2,4,8,19,21,26,32,33],"public":[16,33,36],"return":[10,11,31,32,35],"super":[10,11,23,31,32],"switch":[6,41],"true":[4,10,19,21,32],"try":[4,8,18,20,26,29,32,33,41],"var":43,"while":[1,8,10,26,28,38,41],A:[16,19,23,35,46],As:[1,9,19,20,31,32],At:[10,11,16,32],Being:23,By:[19,25,41],For:[1,4,6,8,10,11,26,27,31,32,33,34,35,36,41,43],IN:1,If:[6,10,16,19,21,33,41,43],In:[1,10,21,29,30,31,32,33,38,41,43,44,46],It:[1,16,18,20,41],Its:[16,41,46],One:[25,41],Or:19,Such:[2,18,19,20,21,41],The:[1,6,8,10,11,18,19,20,21,28,31,32,34,35,38,41,42,46,47],Their:36,Then:[6,21,32],These:38,To:[4,6,8,10,11,26,33,34,38,41],With:[1,2,38,41],__finish:[10,31],__init__:[1,10,11,31,32,41],__main__:[4,8,10,11,26,31,32,34,41],__name__:[4,8,10,11,26,31,32,34,41],_notifi:10,_on_disconnect:10,_on_message_impl:10,a1xfouratnfc9dfqu:[22,41],a925:11,aaai:[23,38],abc:1,abhishek:[41,46],abil:[19,21],abl:[10,19,21],about:[16,19,38],abov:[1,6,18,19,20,21,31,41,43],absolut:21,abstractmethod:1,academia:[41,44],acc:[6,27],acceler:[10,23,29,30,41,44,46],accept:[19,21],access:10,accord:[1,19,32],account:[6,21,22,41,44,46],account_id:[16,19],accur:23,accuraci:[6,11,23,27,28,38],achiev:[10,23],acm:[23,38],across:16,activ:[32,42,43],actual:38,ad:1,adam:38,adapt:[23,27,38],adavantag:10,add:[4,8,19,21,22,26,41],addit:[1,33,43],address:10,adjust:23,admin0000:[18,20],admin:[18,20],adopt:2,advanc:[19,21,23,38,41,43,46],advantag:38,afford:38,after:[1,6,19,21,41,43,46],agent:[10,11,16,19],aggreg:[1,6,9,10,21,23,41],aggregator_dist_adapt:10,agnost:23,ai:[1,4,6,8,16,18,19,20,21,26,27,28,31,32,33,35,36,40,41,43,44,46,47],aim:[38,41,46],aistat:23,alexliang:19,alexqzliang:19,alg:27,algorithm:[1,2,16,19,29,32,36,38,41,46],all:[1,9,10,16,19,21,31,32,38,41,43,46],all_reduc:1,alloc:27,allow:[1,10,38],allreduc:16,almost:31,alogrithm:27,alon:[31,32],along:38,alpha:27,also:[1,2,6,8,11,16,17,19,21,23,26,27,31,32,41,46],although:1,alwai:[4,8,23,26],among:[1,2],amount:38,an:[1,9,21,23,31,38,41,43,46],anaconda3:41,analysi:38,analyt:[16,23],android:[4,5,16,41,44],android_dir:6,angel:[18,20],ani:[1,16,18,19,20,21,35,41,42,43,44,46],annavaram:[41,46],anomali:38,anoth:[1,2,19],answer:[33,36,38],anywher:[16,18,20,21,41,44,46],api:[16,34,44,46],app:[4,27,28],appendix:27,appli:[16,31,32,33,38,41],applic:[1,2,6,16,33,38,41],appreci:[19,21],approach:[23,38],approxif:23,approxim:23,apr:[6,11,31],apt:43,aq:43,ar:[1,9,10,11,16,19,21,28,32,33,35,36,38,40,41,43,46],arbitrari:1,architectur:[2,10,16,23,43,47],area:23,arg:[4,8,10,11,26,31,32,34,35,36,41],argument:[1,4,8,19,21,26,31],arm64v8:43,articl:[23,41,46],arxiv:[23,27,38],ask:[22,40,41],aslo:10,assign:31,assist:[19,21],assum:33,asymmetr:[1,23],asymmetrictopologymanag:1,asymml:23,async_html:0,asynchron:23,attack:[23,36,38],attain:38,attribut:10,author:[41,46],auto:19,autoct:23,autom:[16,23],automat:[10,11,41],automl:[27,28],auxiliari:1,avail:[10,23,31,32,38],averag:[23,39],avestimehr:[25,41,46],avoid:[2,31],await_sync_process_group:10,awar:23,award:[23,25,41,46],b3d7:11,b6c6:10,b6ebbab3ce8a11ecb8af70df029276d3:42,back:[1,16,23,41,47],backdoor:36,backend:[1,10,11,16,31,32,41,46],background:1,bandwidth:[23,29,30,41],bank:[9,41],bar:6,barrier:23,bart:38,base:[1,2,16,23,32,33,34,38,39,41,42,43,44,46],baselin:33,basetopologymanag:1,bash:[6,10,11,31,43],basic:[31,43],basil:23,batch:[6,23,27],batch_siz:[4,10,11,27,28,31,32,34,35,41],becaus:[1,31],beehiv:[6,44],been:[33,38,41,42,46],befor:[1,6,19,21,33],behavior:1,behind:38,believ:38,below:[1,9,10,11,19,21,31,32,33,34,43],benchmark:[16,23,41,46],benefit:[2,23,33,38],bert:38,besid:23,best:[4,8,23,25,26,38,41,46],better:[38,41,44],between:[10,23,33,38],bi:36,bia:32,bin:[6,10,11,31,43],binari:6,bind:21,bit:42,blob:[4,8,17,26,35],blockchain:23,blue:1,boostrap:43,bootstrap:[10,43],both:[38,41,44],bottom:6,bound:[19,21],box:[1,47],brand:6,bridg:38,bring:33,broad:16,broadcast:10,bs:27,buffer:23,build:[1,16,18,20,38,41,42,44],built:[1,2,33,36,38,41,46],busi:33,button:21,byzantin:23,c289:11,c:[0,16,27,42,43],c_opt:27,c_p:27,ca9194865e31:10,call:[1,10,11,32],can:[1,2,6,9,10,17,18,19,20,21,31,32,33,35,36,38,41,43,46],cannot:[1,2],capabl:[16,41,46],capac:38,care:43,career:41,caus:31,cd:[21,43],central:[1,16,18,20,38,41,46],cf:[10,11,17,21,31,32],challeng:[23,38],chang:[11,19,21,34,35,43],chaoyang:[21,25,33,41,43,46],chaoyangh:[21,41,43],chaoyanghe2020fedml:[41,46],charact:[33,36],chart:[10,11],check:[4,8,17,18,19,20,21,26,27,33,36,41,43,47],checkout:43,cheetah:[41,44,46],chmod:43,chosen:27,chrome:0,cifar100:[27,36],cifar10:[6,27,28,36],cifar10_dir:6,cifar:[6,27,33,36],cinic10:[27,36],citi:[16,18,20],clang:43,class_num:[4,10,11,31,32,35],classif:[31,32,35,38],clean:0,cleaningup:10,cleanup_pg:10,clearer:1,cli:[10,11,16,21,27],click:[6,19,21],client1:[10,11],client2:[10,11],client:[6,8,9,10,11,16,21,27,31,32,35,38,41],client_dist_launch:10,client_id_list:[10,11,28,32,34,41],client_index:32,client_login:21,client_manag:[10,11,31],client_master_manag:10,client_model_s3_address:11,client_num:[10,11,31,32,35],client_num_in_tot:[10,11,27,28,31,32,34,41],client_num_per_round:[10,11,27,28,32,34,41],client_numb:35,client_optim:[10,11,27,28,32,34,41],client_silo_config_path:10,client_slave_manag:10,clone:43,cloud:[16,18,20,21,38,41,43,47],cloud_region:[19,21],cluster:[2,23,33],cnn:[2,23,27,36],co:[16,23,41,46],code:[1,2,4,6,8,10,11,16,18,20,21,23,26,31,32,33,35,47],codedprivateml:23,codedreduc:23,collabor:[10,11,18,20,23,38],collect:[23,33,36,38],color:47,com:[1,4,6,8,11,17,18,20,21,22,26,31,32,35,36,40,41,42,43,46,47],come:[15,16,41,43],comm_arg:[10,11,31,32,41],comm_round:[10,11,27,28,32,34,41],command:[10,11,21,31,43],comment:[4,8,26],commerci:38,commit:43,common:[41,46],common_arg:[10,11,32,41],commun:[1,2,16,19,21,23,28,31,32,33,34,38,46],compani:[41,46],compar:[33,38],comparison:[1,2,38],compat:[4,43],competit:38,complet:[1,21,31,32],complex:[10,11,18,19,20,21,41,43],compon:1,comprehens:[33,36,38],compromis:23,comput:[2,16,23,36,41,46,47],computation:38,concern:38,conda:[0,31,32,43],confer:[23,27],config:[4,10,11,17,21,31,41],config_fold:[17,21],config_vers:[19,21],configur:[1,21,31,32,41],congratul:[19,21],connect:[6,10],consid:1,consist:[1,6,16,33,36],consol:[6,21],constrain:38,constraint:2,construct:[1,9],contain:[10,33,36,38,43],contribut:[2,25,41],conv:[27,36],convent:2,convers:1,convinc:33,copi:[19,21,31],copyright:43,core:[2,16,23,38,41,46,47],correct:[23,43],correctli:[6,21],correl:23,correspond:[6,10,11,32],cost:[6,38],counterpart:1,coupl:19,cover:[41,44],cpp:42,cpu:[29,31,32,41],creat:[8,10,11,16,19,26,31,32,34,36,41,42,43],create_mnn_lenet5_model:4,create_model:32,creation:[19,21],credit:43,cross:[4,10,11,16,18,20,21,29,30,31,36,38,39,41,44,46],cross_devic:[4,6,41],cross_silo:[8,10,11,17,21,41],cryptographi:23,cs:6,cubic:23,cuda:43,current:[1,4,6,16,19,21,31,32,41,43],current_running_dir:[19,21],custom:[1,8,16,19,26,41,43,46],custum_data_and_model:[10,11],cut:[16,23,41,44,47],cute:33,cv:27,cvpr:23,cybersecur:[23,38],d7d0:10,d:[27,38],daemon:19,dalla:[18,20],darwin:43,data:[4,8,9,18,20,21,23,26,27,34,36,38,41,44,46],data_arg:[10,11,28,32,41],data_cache_dir:[10,11,31,32,35,41],data_load:[32,35,41],data_silo_select:10,datacent:43,dataload:[4,35],datapath:6,dataset:[2,6,8,10,11,16,26,27,28,29,31,32,34,38,39,41,43,46],dataset_nam:[10,11,31,32,35,41],ddp:[9,10,41],deal:31,debug:[6,21,43],decai:27,decentr:[1,16,18,20,23,33,34,38,41],decentril:39,decomposit:23,deep:[1,23],def:[1,10,11,31,32],defens:23,defici:23,defin:[1,10,31,32],definit:[4,8,26,31,32],degre:38,deliveri:1,demand:[8,26,40,41],demo:[16,18,20,33,41],demonstr:[1,10,33,38],depend:[31,32],deploi:[16,21,33,41],deploy:[16,18,19,20,33,41],deprec:35,describ:[1,19,21],descript:27,deserv:38,design:[1,2,8,16,23,26,31,38,41],desktop:42,despit:38,dest:[17,21],dest_fold:21,destin:21,destination_package_fold:17,detail:[1,2,4,8,10,11,18,20,26,27,32,33,34,41,47],detect:38,determin:[1,35],dev:[19,42],devel:43,develop:[1,2,4,6,8,16,26,29,30,33,38,41,43],devic:[1,4,8,10,11,16,19,21,26,31,32,34,38,39,41,43,44,46],device_arg:[10,11,32,41],device_id:[19,21],df:[17,21],dictionari:35,differ:[1,2,10,16,18,20,31,33,34,36,38,41],differenti:23,difficulti:[10,11],dimens:[32,35],direct:38,directli:[31,38,43],directori:[6,19,21,31,32],dirichlet:27,disconnect:10,discuss:[40,41],dispers:23,displai:16,dist:21,dist_training_arg:10,distribut:[1,2,9,10,11,16,19,21,25,27,38,41,44,46,47],divers:[1,2,16,23,41],dnn:[2,23],doc:[21,41,42,43],docker2:43,document:[16,20,33],doe:[1,16],domain:[38,41],don:33,downgrad:38,download:[6,19,21,31,42],download_mnist:[10,11,31,32,41],dr:[33,41,46],due:[1,2,38],dure:[1,43],e1k05jd1tyw:41,e2b49004db48:10,e798061d62560b03e049d514e7cc8f1a753fde6b:43,e:[1,2,19,21,27,29,30,38,41,43],each:[1,4,6,8,9,10,11,19,21,26,31,32,33,34,35,36,41],earth:[18,20],easi:[18,20],easier:43,easili:2,echo:[31,43],econom:23,ecosystem:[16,37],edg:[6,16,18,19,20,21,23,38,41,44,46,47],edge_case_exampl:36,edge_deploy:21,edge_id:[11,19,21],edit:[32,43],edu:6,ee0b5f53d949c84cee7decbe7a619e63fb1f8408:10,ee0b5f53d949c84cee7decbe7a629e63fb2f8408:[11,32],efa:[29,30,41],effect:38,efficaci:38,effici:[16,18,20,23,38,41],effort:38,egg:41,either:[1,19],elast:23,email:[19,21],embed:[41,44],emnist:[27,33,36,39],empir:38,en:42,enabl:[2,16,18,19,20,23,34,38,41],enable_wandb:[4,10,11,32,41],encapsul:[10,11],end:[10,11,23,31,32,38],engin:[16,18,20,33,41],english:[33,36],enjoi:[19,21],entangl:23,entir:[1,32],entiti:4,entri:[17,21,41,43],entry_point:21,entry_point_fil:17,env:[10,11,31,41,43],environ:[17,31,32,33,38,43],environment_arg:10,ep:[17,21],epoch:[10,11,27,28,31,32,34,41],error:31,errorcod:31,essenti:[2,41],etc:[9,16,19,23,29,30,32,38,41],evalu:[27,38],ex:27,exactli:31,exampl:[1,6,8,16,21,26,31,32,34,35,39,43,46],exce:[22,41],except:10,exchang:1,excit:38,exdb:6,execut:[1,10,31],exist:[1,19,21,33],exit:21,exp:28,expand:41,experi:[1,6,27,33,38,41,44,47],experiment:[16,19,27,28],explor:38,expos:38,expr:31,extend:[33,36],extens:38,facebook:[19,21],facilit:[38,41],fair:[2,23],fairf:23,fairli:33,fals:[4,10,11,32,41],faq:16,far:38,fast:23,fastest:[16,41,44,46],fc:27,feat:[25,41],featur:[4,8,19,21,26],fed_cifar100:[28,36],fed_emnist:36,fed_shakespear:[28,36],fedavg:[1,10,11,27,28,31,32,38,39,41],fedavg_api:32,fedavgclientmanag:31,fedcv:[23,41],feddetect:38,feder:[1,5,10,11,16,18,20,21,25,29,31,32,33,39,41,43,44,46,47],federated_optim:[10,11,28,32,34,41],fedgkt:[33,34],fedgraphnn:[23,41],fediot:41,fedml3:21,fedml:[20,22,23,25,27,28,31,32,34,35,36,37,38,39,44,46,47],fedml_0_0_1:10,fedml_0_0_1_eecbc3ba:10,fedml_0_0_2:10,fedml_0_0_2_8792eda0:10,fedml_0_1_06180ac:11,fedml_0_2_565e48cb:10,fedml_0_2_a33a50ad:11,fedml_aggreg:[6,10,11],fedml_batch_bootstrap:43,fedml_batch_entry_script:43,fedml_client_manag:11,fedml_config:[10,11,31,32,34,41],fedml_cor:1,fedml_data:19,fedml_docker_imag:43,fedml_edge_deploy:19,fedml_main_node_index:43,fedml_main_node_private_ipv4_address:43,fedml_node_index:43,fedml_num_nod:43,fedml_product:21,fedml_run_id:43,fedml_server_manag:10,fedml_startup:43,fedml_torch_fedavg_mnist_lr:[10,11,32],fedml_user_cod:17,fedna:[23,33,34],fednlp:[23,41],fednova:[33,34],fedopt:[33,34,38,39],fedvg:[33,34],feedback:[18,19,20,21],feel:[40,41],femnist:28,few:[8,26,41],field:23,fig:1,figur:[1,9,18,19,20,21,27,33,34,41],file:[4,10,21,31,32,41],fileand:31,filenam:31,fill:[10,11,19,41],financ:16,financi:23,find:[6,21,23,43],finish:[1,10,21,31,32],finsihd:10,first:[1,10,11,31,32,41,42],five:[1,8,26,41],fl:[1,2,4,8,16,19,29,30,31,32,33,34,36,38,41,44,46,47],fl_client:21,flclient_agent_266:21,flexibl:[1,2,8,19,26,31,41],flow:[1,10,11,21,25,31,41],flserver_ag:21,fntml:23,focu:[41,44],focus:1,folder:[6,10,11,19,21],follow:[1,6,8,10,11,16,18,19,20,21,22,26,27,31,32,33,34,35,36,37,39,41,42,43,46],forecast:[16,23],forens:23,forg:0,form:[4,8,26,35],formul:38,forward:[10,11,31,32,38],found:[19,31,41],foundat:[38,41],founder:[16,41,46],four:[18,20,33,36,38],fpcjrkh3bpy:[18,20],fraction:[10,11,31,32],framework:[1,4,10,11,23,26,32,33,34,35,39,41],free:[40,41],frequenc:38,frequency_of_the_test:[10,11,32,41],frequent:[40,41],fri:10,friendli:[1,16,41,44],from:[1,4,8,9,10,11,16,18,19,20,21,23,26,27,28,31,33,34,36,38,41],full:[23,41],fundament:[16,23],funni:[18,20],further:[10,38,47],furthermor:[10,16,38],futur:38,g:[1,2,19,21,29,30,38,41],gap:[2,33,38],gener:[1,31,41,46],generate_topolog:1,geo:19,get:[1,10,16,19,21,32,33,43,46],get_devic:[4,8,10,11,26,31,32,34,41],get_in_neighbor_idx_list:1,get_in_neighbor_weight:1,get_out_neighbor_idx_list:1,get_out_neighbor_weight:1,git:[42,43],github:[1,4,8,11,17,21,26,31,32,35,36,40,41,42,43,46,47],give:[1,31],given:38,global:[9,35,38,41],global_aggregated_model_s3_address:6,global_model:[10,11,41],global_model_file_path:[4,10,11,41],gnn:38,go:6,goal:10,good:33,gpu:[2,9,10,16,29,30,31,33,41,43],gpu_id:[32,41],gpu_map:[10,11,41],gpu_mapping_fil:[10,11,41],gpu_mapping_kei:[10,11,41],grade:[41,44],gradient:23,gradiveq:23,graph:[23,41],green:[6,47],grep:43,group:[10,11,16,18,20,23,27,36,41],grow:38,grpc:[39,41,46],guarante:23,guid:[6,10,11,16,23,42,43],guidanc:41,gz:6,ha:[1,8,10,26,33,38,41,42,46],handl:[1,2,18,19,20,21,33,41],handle_message_receive_model_from_cli:10,handle_message_receive_model_from_serv:10,hang:[41,46],happi:[40,41],have:[1,6,10,16,19,21,31,32,33,41,43],haven:6,havwx1e:[22,41],he:[25,33,41,46],healthcar:16,heck:34,help:[2,16,21,29,32,33,38,41,43],henc:[16,18,20,41],here:[9,10,11,17,31,32,41,43],herein:38,herg:38,hesit:33,hetero:[10,11,28,32,41],heterogen:[23,27],heterosag:23,hierarch:[1,10,16,23,33,34,41],hierarchi:10,hierarchical_fedavg_mnist_lr_exampl:[9,41],high:[16,29,30,38,41,47],highli:[10,11],highlight:1,him:[22,41],hire:[16,41],home:43,hongyi:[41,46],hope:[2,29,32,41],horizont:[16,41],hospit:[9,41],host:[31,33,36,43],hostfil:31,hostnam:31,how:[6,9,10,11,16,18,19,20,31,32,39,41,42],howev:38,html:[0,21,41,42,43],http:[1,4,6,8,11,16,17,18,19,20,21,22,26,27,28,31,32,35,36,38,40,41,42,43,46,47],hub:43,huge:38,human:33,hyper:[19,27,41],hyperparamet:27,i:[31,38,43],iclr:[23,33,34,38],icml:23,id:[1,6,10,11,19,21,27,31,41,43],idea:[29,32,33,41,46],idx1:6,idx3:6,idx:6,ieee:23,ifconfig:10,ignor:1,iid:[27,38],ijcnn:23,ijcv:23,illustr:[1,9,10,11,21,41],imag:[6,10,11,23,31,32,33,36,38,43],imagenet:36,immedi:43,impact:[16,41],implement:[1,2,16,23,27,31,32,41,46],impli:38,improv:38,in_featur:32,in_neighbor_idx_list:1,in_neighbor_weight:1,inc:[16,33,41,43],includ:[1,10,11,21,31,32,33,34,36,38,41,44],increas:38,index:[0,1,10,11,31,32,35],indic:[23,32,38],industri:[16,41,44],infer:[23,41,46],infiniband:[29,30,41],inflex:1,info:[6,10,11,31,32,35,41],inform:[1,23,31,32,41,43,46],infrastructur:33,inher:38,inherit:1,init:[4,8,10,11,26,31,32,34,41],initi:[1,31,32],innov:[1,23],input:[19,21],input_dim:[10,11,31,32],insid:[1,9,10,41],instal:[10,11,16,31,32],instead:31,instruct:[6,19,21,32,33,42,43],insur:16,integ:31,integr:41,interact:43,interest:[16,38,41],interfac:[1,2,10,16,38,41],intern:[16,23,41],internet:[23,41],interpret:43,intrigu:38,introduc:[1,37,38],introduct:[16,41],invit:[10,11,16,22],invok:31,io:[41,42,44],iot:[23,38,41,43,44,46,47],ip:10,ipdp:23,is_mobil:31,isit:23,isregist:19,issu:[42,43],iter:43,its:[1,10,16,38],itself:[1,4,8,26],japan:[18,20],java:43,jetbrain:43,jinhyun:[41,46],join:16,journal:[23,41,46],jsait:23,just:[31,32,33,41],k0nze:42,k:[10,11],kaggl:[33,36],kang:[41,46],kdd:[23,38],keep:[4,8,26,38,41],kei:[10,23,35],kill:[31,43],know:16,knowledg:23,kriz:6,l:31,label:[1,6,23],lack:38,lagrang:23,land:23,landmark:36,languag:[16,23,36,41],laptop:[29,32,33,41],larg:[2,33,38,41,47],last:28,latenc:23,latent:27,later:[19,35],latest:[4,8,26,27,42,43],launch:[19,43],launcher_rdzv_port:10,law:27,layer:[23,35,36],lda:27,leakag:23,learn:[1,2,5,10,11,18,20,21,23,25,29,31,32,33,34,39,41,43,44,46],learning_r:[10,11,28,32,34,41],lecun:6,led:2,left:47,len:6,lending_club_loan:36,length:10,let:[19,21,31,41],level:[1,23,41,46],li:[41,46],lib:41,librari:[1,21,23,25,31,38,42,46,47],licens:43,life:43,lightsecagg:23,lightweight:[16,18,20,23,41,46],like:[2,19,21,41,43,46],limit:23,line:[1,6,8,26,41],linear:[10,11,31,32,36],linearli:23,link:[19,21,27,43],linux:[6,31,41,43,44],list:[1,6,10,19,21,27,33,36,41,43,47],literatur:38,liu:[41,46],live:[20,33,41,46],lkfg3lb13u5:27,ll:37,lo:[18,20],load:[4,8,10,11,26,31,32,34,35,36,41],load_cross_silo:10,load_data:[10,11,31,32,35,41],load_for_simul:32,load_partition_data_mnist:[10,11,31,32],load_synthetic_data:41,loader:[10,11,16,19,31,32],local:[9,16,19,35,38,41],local_log_output_path:10,local_serv:21,local_test_on_all_cli:32,locat:[6,10,11,21,35,43],log:[10,11,19,21,31,32,41],log_file_dir:[10,11,19,21,32,41],logger:[10,11,31,32,35],login:[10,11,16],logist:[27,29,32,33,36,41,43],logisticregress:[10,11,31,32],logo:41,logout:16,look:[31,43],loss:[6,10,11,31,32],low:[1,23,41,46],lower:[33,36],lr:[10,11,27,28,32,41],lstm:[27,36],m:32,machin:[1,2,16,18,20,23,32,41,44,46],maco:[6,19,21,43],mai:[1,10,19,21,31,33,38,41,43],main:[31,32,41,43],maintain:[16,38,43],make:[0,1,32,38,43],manag:[10,11,16,41,46],mani:27,mapping_default:[10,11,41],mapping_serv:10,mapping_silo_1:10,mapping_silo_2:10,mar:43,march:[41,46],market:23,massiv:38,master:[1,4,8,10,11,17,21,26,31,32,35,36,41,43,46],master_address:10,match:10,matrix:23,maxdepth:39,maximum:[16,18,20,41],mb2:19,mbp:[21,43],md:[39,43],mean:[6,9,19,31,32,41,43],meaning:38,mechan:1,meet:[1,8,26,41],member:[19,21,22,41],memlock:43,memori:[2,23,38],merkl:23,messag:[10,11,16,21,41],message_json:[6,11],message_kei:[10,11],messeng:[19,21],mest:23,met:[42,43],method:[1,23,27,31,33,41],metric:27,mi:[41,46],microsoft:42,migrat:21,milena:23,mimick:33,mine:[16,23],minist:[31,32],minut:41,mission:[16,41],mitig:[23,33,38],mix:23,mkdir:6,ml:[23,25,27,33,41],mlop:[33,44,46],mlops_metr:[6,11],mlsp:23,mlsy:[23,25,41],mnist:[6,27,29,31,32,33,35,36,39,41,43],mnist_dir:6,mnistdataset:4,mnn37:[21,41,43],mnn:4,mobil:[4,47],mobilenet:[27,36],mode:[6,31,43],model:[2,4,8,18,19,20,21,23,26,28,29,34,38,41,43,44,46],model_arg:[10,11,28,32,41],model_dict:6,model_file_cach:[10,11,41],model_file_cache_fold:[10,11,41],model_hub:32,model_nam:32,model_train:32,modern:2,modifi:[11,19,31],modul:[10,11,31,32,41,46],monitor:[10,11,19,21],more:[1,2,10,11,16,19,20,21,32,33,34,35,38,43],moreov:38,most:[1,38],move:6,mpi4pi:[1,31,42],mpi:[1,16,33,39,41,44,46],mpi_abort:31,mpi_base_framework_exampl:31,mpi_comm_world:31,mpi_decentralized_fl_exampl:31,mpi_host_fil:31,mpi_torch_fedavg_mnist_lr_exampl:[29,31,41,43],mpirun:31,mqtt:[10,11,39,41,46],mqtt_config:[10,11,41],mqtt_config_path:[10,11,41],mqtt_s3:[10,11,41],mqtt_s3_comm_manag:6,mqtt_s3_fedavg_hierarchical_mnist_lr_exampl:10,mqtt_s3_fedavg_mnist_lr_exampl:[6,9,11,21,41],mqtt_s3_multi_clients_comm_manag:[10,11],mqtt_s3_status_manag:10,msclkid:42,msg:[10,11],msmpi:42,multi:23,multipl:[9,10,23,38,41,43],multiprocess:1,murali:[41,46],mv:6,my:[6,19,21],my_dataset:4,my_model_trainer_classif:[10,11,31,32],mymodeltrain:32,myst:0,mysteri:38,n:[1,32],n_dist_train:10,n_node_in_silo:10,n_proc_in_silo:10,n_proc_per_nod:10,na:23,naacl:[23,38],name:[4,19,21,38,39,42,43],namespac:[19,21],nativ:43,natur:[16,23,36,41],nccl:[16,41,44,46],necessari:[1,38],necessit:38,need:[1,2,10,11,16,18,19,20,21,31,32,33,38,41,43],neighbor:1,neighbor_num:1,neighborhood:1,nerdschalk:42,net:[23,27,43],network:[1,23,41],network_interfac:10,neural:[2,23,41,46],neurip:[23,25,33,34,41,46],newli:[4,8,19,21,26,33],newton:23,next:[6,21,33,36,43],nlp:[27,38],nn:[1,10,11,31,32],node:[1,10,16,18,20,41],node_address:10,node_index:1,nofil:43,non:[2,27,31,33,38],none:[4,6,10,21,31],nonetyp:10,normal:[1,27,35,36,43],note:[1,10,11,17,19,21,27,28,31,43],notifi:10,novel:2,now:[19,20,21,31,32,41,43,46],np:31,number:[1,6,8,10,11,19,26,27,31,35,38,41],nus_wid:36,nvidia:43,nvlink:[29,30,41],nwp:36,object:[10,31,32,38],obtain:38,occur:38,octopu:44,offici:31,omnilyt:23,on_disconnect:10,on_messag:10,onc:[21,32],one:[8,26,31,32,41],one_lin:[10,11,21],ones:[19,38],onli:[1,2,4,8,21,26,31,41],open:[0,6,16,18,19,20,21,23,31,33,38,44,47],openreview:27,oper:[4,6,31,32,41,42,44,46],opportun:23,opt:41,optim:[2,23,27,38,39],option:[6,21],orchestr:16,order:[1,10],org:[1,27,38],organ:[16,38,40,41,44,46],orient:[2,33,36],origin:[10,11,31,32,33,36,41,43,46],os:43,os_nam:[19,21],other:[1,19,21,31,32,33,41,43,44,46],our:[1,2,4,8,10,11,16,18,19,20,21,26,27,31,33,38,41,42],out:[1,2],out_directed_neighbor:1,out_featur:32,out_neighbor_idx_list:1,out_neighbor_weight:1,output:[6,10,11,21,31,32,35,41],output_dim:[8,10,11,26,31,32,34,41],over:[1,16,23,31,38],overdesign:31,overflow:[33,36],overhead:38,overview_training_ori:1,own:[1,16,19,21,43],p:6,pachinko:27,pack:10,packag:[16,19,31,32,41,43,46],page28:27,page34:27,page:[6,19,21,27],paper:[10,11,16,25,27,28,31,32,36,38,41,46,47],paradigm:[2,38],paradistributeddataparallel:1,paragraph:27,parallel:[1,10],param:[10,35],paramet:[1,19,27,38,41],parrot:[8,34,35,44,46],pars:19,parser:0,part:[1,31,32],partial:23,particip:1,particularli:38,partit:[19,27,38],partition_alpha:[10,11,28,32,41],partition_method:[10,11,28,32,41],pass:[1,16,21,41],password:[18,20],passwrodless:10,path:[1,10,19,21],patit:27,patition_alpha:27,pattern:[1,2,31],payload:31,pdf:[27,38],pdsh:10,peilin:[41,46],peopl:16,per:38,perform:[10,11,19,21,23,33,36,38,39],permiss:43,person:[18,20,23,29,32,38,41],perspect:1,pet:33,pg:10,phase:43,philosophi:[1,8,26,31,41],pi:[38,43],pick:19,pip:[19,31,32,33,41,42],pipe:23,pipelin:[23,41,44],pipetransform:23,plan:11,platform:[16,18,19,20,33,38,41,46],pleas:[4,6,8,10,11,16,18,19,20,21,22,26,27,31,32,33,34,35,36,40,41,42,43,47],plugin:33,point:[2,21,38,41],polynomi:23,polyshard:23,popular:[1,38],port:10,possibl:[8,26,38,41],post:[42,43],pow:27,power:27,practic:2,praneeth:[41,46],prcoess:10,predefin:1,predict:[23,33,36],prefer:19,prepar:[6,20,43],preprocess:38,present:[10,31,32,38,41,46],preserv:[23,38],press:21,print:1,privaci:[1,16,18,19,20,38,41,46],privat:[18,20,23],private_config_path:10,privileg:43,problem:23,proce:31,procedur:1,process:[1,10,16,19,21,23,31,33,34,35,36,39,41,44,46],process_num:31,processor:42,product:[16,17,38,41],professor:33,profil:[10,11],program:[2,31,32,43],progress:[19,21,38],prohibit:38,project:[1,4,10,11,16,41,42,43],promis:38,prompt:43,propos:38,protect:23,protocol:1,provabl:[16,18,20,41],provid:[1,10,11,16,18,19,20,21,32,33,38,41,44,46],ps:43,pt:[10,11,41],publicli:38,publish:[33,38],pull:43,purpos:[41,46],push:[6,38],py3:41,py:[4,6,8,10,11,17,21,26,31,32,35,41,43],pycharm:43,pypi:1,python3:[10,11,41],python:[4,6,8,10,11,16,21,26,31,32,35,36,41,42,43,44,46],pytorch:[1,9,10,35,39,41,46],q:43,qiang:[41,46],qiu:[41,46],qr:6,quadrat:23,quantiz:23,queri:1,question:[33,36,38,40,41,42],quick:[29,32,33],quickli:[31,32],r34w:[22,41],r:[27,31],ramesh:[41,46],random_se:[10,11,32,41],rang:[16,38],rank:[10,11,31,41],rapidli:38,rare:38,raskar:[41,46],raspberri:[38,43],rate:[23,27,38],rather:1,read:[10,11,31,32,33,35,41],readi:[10,19,21],readm:43,real:[10,11,21,23,27,31,33,38,41,44],realist:[2,27,38],realli:36,receiv:[1,9,10,41],receive_id:10,recent:38,recommend:[10,11],reduc:[8,10,11,26,41],ref:[1,28],refactor:[6,35,36,41,43,46],refer:[1,6,10,11,16,27,28,32,38,43],referenc:28,reference_exampl:1,reflect:[1,41,43],reformul:23,refresh:21,regist:21,register_message_receive_handl:1,registr:21,regress:[27,29,32,33,36,41,43],regul:38,rel:19,relat:[1,4,16,19,21,27,31,32,33,41,46],releas:[6,17,19,21],release_401:43,relev:6,rememb:[6,43],remot:[1,43],remov:35,rendezv:10,reorgan:[41,46],replac:21,report:[27,31],report_aggregated_model_info:6,report_client_model_info:11,report_server_training_metr:[10,11],report_server_training_round_info:11,repres:[23,31,33,34,36,38,41,47],represent:38,reproduc:27,requir:[1,6,43],research:[1,2,16,18,20,29,31,32,33,36,38,41,44,46,47],resili:23,resnet18:[27,36],resnet18_gn:28,resnet56:28,resnet:[27,29,32,36,41],resolut:23,resourc:[4,8,26,33,38,46],respect:10,respons:1,restart:43,restrict:[1,38],result:[1,9,19,21,33,38,41],reus:[2,47],review:[19,21,23],rf:6,rich:33,right:6,ring:1,rm:[6,43],rnn:[27,36],robust:[23,39],root:32,round:[6,10,11,23,27,28,31,32,38],round_id:[10,11,31],round_idx:[6,11],round_index:11,round_numb:10,rpc:[1,41,46],rtx2080tix4:43,run:[0,4,6,8,9,10,11,16,19,26,28,31,32,33,34,41,44],run_client:[10,11],run_cross_silo_cli:11,run_cross_silo_serv:11,run_hierarchical_cross_silo_cli:10,run_hierarchical_cross_silo_serv:10,run_id:[6,11,41],run_nam:32,run_one_line_exampl:[31,43],run_serv:[6,10,11],run_simul:[8,26,31,32,41],run_step_by_step_exampl:31,running_tim:11,s3:[10,11,39],s3_config:[10,11,41],s3_config_path:[10,11,41],s:[10,11,19,21,23,31,32,35,41,44,47],salman:[25,41,46],same:[1,10,31,32,43],sampl:[10,11,31,32,35],saniti:[17,33],scalabl:[23,25,41],scale:[2,16,18,20,33,38,41,44,46,47],scenario:[2,10,16,41,46],schedul:[23,38],scientif:[16,41],scm:42,scope:16,screenshot:[19,21],script:[6,10,11,19,21,43],sdcard:6,sdk:[4,16,41,44,47],seamlessli:[16,21],search:23,sec:1,second:1,section:[1,16,27,31,32],secur:[1,16,18,20,38,41,46],sed:31,see:[1,6,10,11,18,19,20,21,31,32,37,43],segment:38,select:[19,21,23],self:[1,6,10,11,23,31,32],semi:23,seminar:[25,41],send:[1,10,11,19,21,31],send_messag:[1,10,11],send_message_sync_model_to_cli:10,sens:41,sensi:[23,38],sensor:38,sent:[10,11],separ:[1,2],seq2seq:38,sequenc:[23,38],seri:[16,23,38],serial:[6,23],serv:[10,11,21,23,33,41,44,46],server:[4,8,9,10,11,16,21,29,31,32,41,43],server_config_path:10,server_gpu_mapping_kei:10,server_manag:10,serverless:23,servermnn:4,set:[1,10,19,27,31,32,38,41,43],setup:43,sf:[17,21],sgd:[1,10,11,23,27,28,32,34,41],sh:[0,6,10,11,17,31,43],shakespear:[27,29,32,33,36,41],shallow:[10,11,31,32],shard:23,share:[38,41,46],shared_invit:[22,41],shen:[41,46],shm:43,should:[1,6,10,19,21,31,32,43],show:[1,6,18,19,20,21,39,41],showcas:38,shown:[1,9,10,11,19,21,31,32,33,34],shuffl:4,side:[10,11,47],sigmoid:[10,11,31,32],sign:21,signal:23,signific:38,silo:[9,10,11,18,20,21,36,39,41,44,46],silo_1:10,silo_2:10,similar:[1,6,31,32],simpl:[21,34,39,41,44],simplest:[1,35],simplic:[8,26,41],simplifi:[16,18,20,33],simul:[2,26,33,34,39,41,43,44,46],simulation_sp:41,simulatormpi:[26,31,34],simulatorsingleprocess:34,simultan:[8,23,26,41],singh:[41,46],singl:[10,16,31,33,34,39,41,44,46],single_process:41,site:41,size:[10,11,27,31,32,43],skeleton:33,slack:[16,19,21,41],slave:10,small:[23,29,32,41],smart:16,smartphon:[5,41,44,46],smoothli:[33,41],snippet:1,so:[32,41,46],social:[18,20],sock:43,softwar:[31,47],solut:[16,38],solv:[23,38],some:[1,38,41,43,46],someth:43,songz:[41,46],soon:[6,15,19,21,43],sourc:[1,4,6,8,10,11,16,17,21,26,35,44,47],source_fold:[17,21],sourcecod:[21,43],sp:32,sp_fedavg_mnist_lr_exampl:[29,32,41],spars:23,sparsiti:38,speak:33,special:[40,41],specif:[1,2,6,10,16,19,21,31,32,38,41,43],specifi:[6,10,21,31,32,34,43],sphinx:0,spider:23,split:[1,10,19,33,34,38],split_data_for_dist_train:10,spreadgnn:23,ssfl:23,ssh:10,stack:[23,33,36,41],stackoverflow:[27,36,42],stackoverflow_lr:28,stackoverflow_nwp:28,stage:16,stand:[16,27,31,32],standalon:[1,2,16,31,41],standard:1,stanford:[25,41],start:[4,8,10,11,16,19,26,31,32,33,34,46],start_train:21,starter:41,statist:2,statu:[10,11,19,21],step1:[31,32],step2:[31,32],step:[6,19,21],still:[31,32],stop:10,stop_receive_messag:10,stop_train:21,storag:38,str:1,straggler:23,strategi:38,structur:[35,38],student:33,studi:38,studio:43,submit:2,subscrib:21,subsequ:31,successfulli:[6,19,21,31,32],sudo:43,suggest:[2,29,30,38,41],suitabl:38,summar:27,summari:43,sun:41,supervis:23,support:[1,4,16,19,21,33,34,36,38,40,41,43,44,46],sure:32,swap:19,symmetr:1,symmetrictopologymanag:1,sync_process_group:10,synchron:1,synthet:[27,29,32,33,36,41],system:[1,2,10,11,16,19,21,33,41,42,46],systemat:[38,43],systemctl:43,t:[6,17,21,22,33,41,43],table1:27,table2:27,table7:27,table8:27,tacc:23,tackl:23,tag:[33,36,38,43],tail:36,tailor:[36,41,46],take:[10,35],talk:25,tangibl:33,tar:6,target:23,task:[23,27,32,33,35,36,43],teach:33,team:33,technolog:[16,23,41],tediou:[33,43],templat:[41,46],tencent:38,tensorflow:1,termin:[0,31],test:[6,10,11,17,28,31,32,35,42],test_acc:[11,32],test_data_glob:[10,11,31,32,35],test_data_local_dict:[10,11,31,32,35],test_data_num:[10,11,31,32,35],test_dataload:4,test_dataset:4,test_loss:[11,32],test_on_server_for_all_cli:[6,10,11],test_path:[10,11,31,32,35],text:[21,23,38],textit:1,texttt:1,than:[1,16,38],thank:[16,38,41],thei:43,them:[10,31],theoret:1,theori:23,therefor:10,thi:[1,2,6,10,11,16,17,19,21,29,30,31,32,33,36,38,39,41,42,43,46,47],thing:[23,41],though:[31,32],three:38,through:[1,6,10,11,21,31,32,35],thu:[1,6],tier:[27,33,36],time:[2,6,16,23,31,38],timestamp:11,titl:[41,46],tnnl:23,toctre:39,tokyo:[18,20],tool:[19,21,42],toolkit:43,top:[6,27,33,36],topic:10,topolog:[19,21,23,33,34,41,46],topologymanag:1,torch:[1,10,11,31,32],torch_client:[10,11,17,21],torch_fedavg_mnist_lr_custum_data_and_model_exampl:[31,32],torch_fedavg_mnist_lr_one_line_exampl:[31,32],torch_fedavg_mnist_lr_step_by_step_exampl:[31,32],torch_serv:[10,11,17,21],toronto:6,total:[10,31,35],total_round:11,tour:41,toward:23,tpmgr:1,track:[10,11,16],tracking_arg:[10,11,32,41],tradeoff:23,train:[1,2,4,6,8,9,18,19,20,26,34,35,38,41,44,46,47],train_arg:[10,11,28,32,34,41],train_data_glob:[10,11,31,32,35],train_data_local_dict:[10,11,31,32,35],train_data_local_num_dict:[10,11,31,32,35],train_data_num:[10,11,31,32,35],train_dataload:4,train_dataset:4,train_path:[10,11,31,32,35],trainer:[10,41,46],trainer_dist_adapt:10,training_acc:[10,11,32],training_loss:[10,11,32],training_num:6,training_typ:[10,11,32,41],transact:23,transfer:23,transform:[2,23,38],transplant:[16,33],tree:[1,11,21,23,31,32,35,36,41,46],trend:38,trick:38,trivial:[2,33],trpc:39,trustworthi:[25,41],tue:31,tune:41,turbo:23,turn:6,tutori:[16,19,39,41,42],two:[1,6,10,31,32,33,36,43],type:[1,10,11,19,21,31,38,43],u:43,ubuntu20:43,ubyt:6,uci:36,ui:[18,19,20,21,41],ulimit:43,under:[9,19,20,21,23,31,32,35,38],underli:1,understand:[2,38],undirect:1,undirected_neighbor_num:1,unifi:38,uniformli:[10,11,31,32],unique_device_id:[19,21],univers:38,unravel:38,unstructur:23,up:21,updat:[10,11,31,32,43],upgrad:[32,33,41,43,46],upload:[16,19,31],upper:[33,36],url:[19,21],us:[1,2,4,6,10,11,16,19,21,23,29,30,31,32,33,34,35,36,38,39,41,42,43,44,46],usa:[18,20],usag:[21,36,43],usb:6,user:[1,6,16,18,20,31,38,42,44,46],userid:[17,19],usernam:[18,20],using_gpu:[10,11,32,41],using_mlop:[10,41],usr:[10,11,31],util:[1,19,21,38,41,46],v:[17,18,20,41,43,46],valid:28,validation_arg:[10,11,32,41],valu:[19,21,32,33,35],valuabl:[2,19,21],variabl:43,variou:[1,16,18,20,38,41],vector:23,vepakomma:[41,46],verifi:23,versatil:[23,41,44],version:[1,2,16,19,21,31,32,41,43,46],vertic:[1,16,33,34],vfl:36,via:[16,19,23,33,36],video:[16,18,20],vision:[16,36,41,47],visionari:[23,41],visit:[18,20,27,33],visual:[10,11,19,21,42,43],visualstudio:[42,43],vldb:23,vmlldzoxode2ntu:27,wa:31,wai:31,wait:10,wandb:[4,27,28],wandb_ent:32,wandb_kei:[10,11,32],wandb_nam:[10,11],wandb_obj:4,wandb_project:[10,11,32],wang:[41,46],want:43,watch:[18,20,41,46],wd:27,we:[1,2,4,6,8,10,11,19,21,23,26,27,29,30,31,32,37,38,40,41,43,44],web:[6,18,19,20,21,41],websit:[33,36],wechat:[16,41],wed:11,weight:1,weight_decai:[10,11,32,34,41],welcom:[18,20],well:[6,33,47],wget:6,what:31,when:[1,21,29,31,32,35,41,43],where:[10,43],whether:43,which:[1,2,4,9,10,21,31,32,33,41,46],white:[27,41,46],whole:31,wide:[1,38],wider:38,william:[33,36],win:42,window:[6,10,11,31],wireless:23,wise:23,without:[1,2,16,18,19,20,21,33,35,41,43],won:[41,46],word:[33,36,41,44],work:[1,6,16,19,33,36,38],workabl:31,worker:[2,10],worker_num:[11,31,41],workermanag:1,workflow:[18,19,20,21,41],workshop:[23,25,38,41,46],workspac:43,world:[10,11,18,20,21,23,25,33,38,41,44],wors:38,would:[19,21,43],wrapper:41,write:[21,31],www:[6,18,20,41,43,46],x64:42,x86_64:43,x:[10,11,31,32],xda:6,xiaoyang:[41,46],xxx:[6,32],xzvf:6,y:43,yaml:[10,11,31,32,34,41],yaml_config_fil:41,yaml_path:41,yan:[41,46],yang:[41,46],yann:6,ybv29kak:28,ye:36,year:[16,23,33,41,46,47],yellow:1,you:[6,10,11,16,17,19,21,22,31,32,33,36,37,39,41,42,43],your:[16,19,21,29,30,32,40,41,43],youtub:[18,20,41,46],zero:[16,18,20,41],zhang:[41,46],zhao:[41,46],zip:[21,32,41],zoo:[41,46],zt:[22,41]},titles:["Install","FedML APIs (core)","FedML APIs (high-level)","<no title>","FedML BeeHive API Reference","FedML Beehive Examples","Federated Learning on Android Smartphones","FedML Beehive User Guide","FedML Octopus API Reference","FedML Octopus Examples","FedML Octopus Example with MNIST + Logistic Regression","FedML Octopus Example with MNIST + Logistic Regression","FedML Octopus User Guide","FedML Cheetah API Reference","Examples","FedML Cheetah User Guide","Welcome to FedML","FedML MLOps CLI and API Reference","FedML MLOps Live Tutorial","FedML MLOps User Guide","Tutorials for MLOps","FedML MLOps User Guide","Join the Community","Research Publications","Slides","Video","FedML Parrot API Reference","Benchmarking Results for MPI-based federated learning","Benchmarking results when using single process-based simulation","FedML Parrot Examples","Simulation with NCCL-based MPI (the fastest training)","Simulation with Message Passing Interface (MPI)","Simulation with a Single Process (Standalone)","FedML Parrot User Guide","Algorithm Reference Implementations","Customizing Data Loader for Your Own Dataset","Datasets and Models","Concepts","Ecosystem","<no title>","FedML FAQ","Get Started","FedML Installation on Windows OS","Installing FedML","Mission","FedML MLOps Video Tutorial","Open Source Code Architecture","Technology Overview"],titleterms:{"1":[6,17,19,21,31,32,47],"10":28,"2":[6,17,19,21,31,32,47],"3":[6,17,19,21,31,32,47],"4":[17,19,21,31,32],"5":21,"56":28,"new":31,"public":[23,41,47],A:[7,10,11,38,41,47],One:[10,11,31,32],account:19,account_id:21,adb:6,agent:21,ai:[10,11,17,23],algorithm:[23,31,33,34,47],allreduc:12,an:19,anaconda:43,android:[6,7,37,43],api:[1,2,4,8,10,11,13,17,26,31,32,36,41],apk:6,app:43,applic:[7,21,23,47],architectur:[41,46],autom:21,backend:12,base:[7,12,27,28,29,30,31],basic:37,beehiv:[4,5,7,16,41],benchmark:[27,28,33,38],better:[10,11],beyond:1,bind:6,build:[0,17,21],c:7,career:16,cheetah:[13,15,16],cifar100:28,cifar:28,cli:17,client:[7,17,19],cnn:28,code:[12,19,41,42,43,46],collabor:[16,19,21,47],command:6,commun:[12,22,41],comput:38,concept:37,conda:42,configur:19,connector:16,coordin:1,core:1,creat:21,cross:12,custom:[10,11,31,32,33,35],cv:36,data:[6,10,11,16,19,31,32,33,35],dataset:[33,35,36],decentr:31,definit:1,demo:7,deploi:[0,7],deploy:[21,47],design:47,develop:[21,31],devic:[6,12],differ:12,displai:17,distribut:[12,23],divers:[12,38],dnn:27,docker:[42,43],doe:21,download:36,ecosystem:[38,41],edit:19,emnist:28,engin:7,exampl:[5,9,10,11,14,17,29,33,41],exampless:41,experi:[10,11,19],experiment:21,faq:[40,41],fastest:[29,30],featur:41,fed:28,fedcv:38,feder:[6,9,12,23,27,28,36,37,38],fedgraphnn:38,fediot:38,fedml:[1,2,4,5,6,7,8,9,10,11,12,13,15,16,17,18,19,21,26,29,33,40,41,42,43,45],fedna:36,fednlp:38,financ:36,five:[10,11],fl:[6,7,9,12,23],framework:[31,38],friendli:47,from:[17,42,43],full:47,get:41,googl:28,gradient:1,graph:38,group:[19,21,22],guid:[7,12,15,19,21,33,41],guidanc:43,hierarch:[9,12],high:[2,23],horizont:[9,36],how:21,hyper:28,iid:28,impact:[23,47],implement:[33,34],industri:47,instal:[0,6,21,33,41,42,43],interfac:[29,31],internet:38,introduct:12,invit:[19,21,41],iot:16,join:[22,41],land:16,languag:38,laptop:6,larg:[16,23],leaf:28,learn:[6,9,12,16,27,36,37,38],level:[2,47],librari:[37,41],lightweight:27,line:[10,11,31,32],linear:27,live:18,loader:[33,35],local:[12,21],login:[17,19,21],logist:[10,11,28],logout:17,manag:[1,19,21],messag:[1,29,31],method:38,mission:44,ml:47,mlop:[6,7,10,11,16,17,18,19,20,21,37,41,45,47],mnist:[10,11,28],mobil:6,model:[1,10,11,16,27,31,32,33,36],modern:27,more:41,mpi:[27,29,30,31,42],natur:38,nccl:[29,30],network:[27,38],neural:[27,38],nlp:36,non:28,octopu:[8,9,10,11,12,16,41],open:[10,11,17,37,41,46],optim:47,option:19,orchestr:7,orient:1,os:42,other:36,our:22,outlin:[16,23,33],overview:[16,17,41,47],own:[33,35],packag:[17,21],paper:23,paramet:[28,31,32],parrot:[12,16,26,29,33,41],pass:[29,31],path:6,pip:43,platform:[6,17,21,47],prepar:[31,32],preview:0,privaci:[23,47],privat:19,process:[28,29,32,38],program:1,project:[19,21],provid:36,python:7,quick:41,real:[12,16,47],realiti:16,recommend:43,recruit:41,refer:[4,8,13,17,26,33,34,41,46],regist:19,regress:[10,11,28],repositori:43,reproduc:28,research:23,resnet:28,resourc:[16,41],result:[10,11,27,28,31,32],rnn:28,run:[21,43],scale:[21,23],scenario:12,scientif:[23,47],sdk:[7,43],seamlessli:12,secur:[23,47],server:[6,7,17,19],servic:36,set:6,setup:[31,32],shakespear:28,shallow:27,silo:[12,16],simpl:16,simplifi:[21,41,47],simul:[12,16,28,29,30,31,32],singl:[28,29,32],slack:22,slide:24,smartphon:[6,16],sourc:[19,37,41,42,43,46],speedi:16,split:36,stack:47,stackoverflow:28,standalon:[29,32],start:[6,21,41],step:[31,32],succeed:43,support:12,synthet:[6,19],system:[23,38,47],talk:41,task:38,technolog:47,test:43,thing:38,tool:6,topolog:1,track:[19,21],train:[7,10,11,12,16,21,23,29,30,31,32],trainer:1,transfer:6,transplant:12,troubleshoot:43,tutori:[18,20,45],updat:19,upload:21,us:[27,28],user:[7,10,11,12,15,19,21,33,41,43,47],version:17,vertic:[9,36],via:21,video:[25,41,45,46],view:[31,32],vision:[23,38],visionari:47,wechat:22,welcom:16,when:28,why:16,window:[42,43],work:21,worker:1,world:[12,16,47],your:[6,33,35]}})
\ No newline at end of file
diff --git a/doc/en/_static/image/MLOps_workflow.png b/doc/en/_static/image/MLOps_workflow.png
new file mode 100644
index 0000000000..04d39bb4d1
Binary files /dev/null and b/doc/en/_static/image/MLOps_workflow.png differ
diff --git a/doc/en/_static/image/mlops_workflow.png b/doc/en/_static/image/mlops_workflow.png
deleted file mode 100644
index d40834697e..0000000000
Binary files a/doc/en/_static/image/mlops_workflow.png and /dev/null differ
diff --git a/doc/en/_static/image/mlops_workflow_new.png b/doc/en/_static/image/mlops_workflow_new.png
new file mode 100644
index 0000000000..cd45a05547
Binary files /dev/null and b/doc/en/_static/image/mlops_workflow_new.png differ
diff --git a/doc/en/mlops/user_guide.md b/doc/en/mlops/user_guide.md
index f3d95685ed..48578bacf7 100644
--- a/doc/en/mlops/user_guide.md
+++ b/doc/en/mlops/user_guide.md
@@ -1,9 +1,9 @@
# FedML MLOps User Guide
-**Write Once, Run Anywhere: Seamlessly Migrate Your Local Simulation to the Real-world Edge-cloud Deployment**
+**Write Once, Run Anywhere: Seamlessly Migrate Your Local Development to the Real-world Edge-cloud Deployment**
- How Does FedML MLOps Platform Work?
-- Local Development (Simulation) and Building MLOps Packages
+- Local Development and Building MLOps Packages
- Create Application and Upload Local Packages
- Install FedML Agent: fedml login $account_id
- Invite Collaborators, Create a Group and a Project
@@ -12,7 +12,7 @@
### How Does FedML MLOps Platform Work?
- \
+ \
Figure 1: the workflow describing how MLOps works
MLOps works as the figure shown above. Such a workflow is handled by web UI without the need to handle complex deployment.
@@ -22,7 +22,7 @@ Now please register an account and follow the following instruction step by step
Please open [https://open.fedml.ai](https://open.fedml.ai) and find "Sign Up".
-### Local Development (Simulation) and Building MLOps Packages
+### Local Development and Building MLOps Packages
After you finish the local developing/debugging of the FedML project using FedML library (e.g., successfully run the example [https://doc.fedml.ai/cross-silo/examples/mqtt_s3_fedavg_mnist_lr_example.html](https://doc.fedml.ai/cross-silo/examples/mqtt_s3_fedavg_mnist_lr_example.html)),
you can now deploy it into the real-world edge-cloud system.
diff --git a/research/Awesome-Federated-Learning.md b/research/Awesome-Federated-Learning.md
new file mode 100644
index 0000000000..77332a78e7
--- /dev/null
+++ b/research/Awesome-Federated-Learning.md
@@ -0,0 +1,1412 @@
+## FedML - The Most Popular Federated Learning Library https://fedml.ai
+
+# Awesome-Federated-Learning [](https://awesome.re)
+
+A curated list of federated learning publications, re-organized from Arxiv (mostly).
+
+Last Update: July, 20th, 2021.
+
+If your publication is not included here, please email to chaoyanghe.com@gmail.com
+
+# Foundations and Trends in Machine Learning
+We are thrilled to share that [Advances and Open Problems in Federated Learning](https://arxiv.org/abs/1912.04977) has been accepted to [FnTML](https://www.nowpublishers.com/MAL) (Foundations and Trends in Machine Learning, the chief editor is [Michael Jordan](https://people.eecs.berkeley.edu/~jordan/)).
+
+[A Field Guide to Federated Optimization](https://arxiv.org/abs/2107.06917)
+
+
+## Publications in Top-tier ML/CV/NLP/DM Conference (ICML, NeurIPS, ICLR, CVPR, ACL, AAAI, KDD)
+### ICML
+| Title | Team/Authors | Venue and Year | Targeting Problem | Method |
+|---|---|---|---|---|
+| [Federated Learning with Only Positive Labels](https://arxiv.org/pdf/2004.10342.pdf) | Google Research | ICML 2020 | label deficiency in multi-class classification | regularization |
+| [SCAFFOLD: Stochastic Controlled Averaging for Federated Learning](https://arxiv.org/abs/1910.06378) | EPFL, Google Research | ICML 2020 | heterogeneous data (non-I.I.D) | nonconvex/convex optimization with variance reduction |
+| [FedBoost: A Communication-Efficient Algorithm for Federated Learning](https://proceedings.icml.cc/static/paper_files/icml/2020/5967-Paper.pdf) | Google Research, NYU | ICML 2020 | communication cost | ensemble algorithm |
+| [FetchSGD: Communication-Efficient Federated Learning with Sketching](https://arxiv.org/abs/2007.07682) | UC Berkeley, JHU, Amazon | ICML 2020 | communication cost | compress model updates with Count Sketch |
+| [From Local SGD to Local Fixed-Point Methods for Federated Learning](https://arxiv.org/pdf/2004.01442.pdf) | KAUST | ICML 2020 | communication cost | Optimization |
+
+### NeurIPS
+| Title | Team/Authors | Venue and Year | Targeting Problem | Method |
+|---|---|---|---|---|
+| Lower Bounds and Optimal Algorithms for Personalized Federated Learning | KAUST | NeurIPS 2020 | non-I.I.D, personalization | |
+| Personalized Federated Learning with Moreau Envelopes | The University of Sydney | NeurIPS 2020 | non-I.I.D, personalization | |
+| Personalized Federated Learning with Theoretical Guarantees: A Model-Agnostic Meta-Learning Approach | MIT | NeurIPS 2020 | non-I.I.D, personalization | |
+| Differentially-Private Federated Contextual Bandits | MIT | NeurIPS 2020 | Contextual Bandits | |
+| Federated Principal Component Analysis | Cambridge | NeurIPS 2020 | PCA | |
+| FedSplit: an algorithmic framework for fast federated optimization | UCB | NeurIPS 2020 | Acceleration | |
+| Federated Bayesian Optimization via Thompson Sampling | MIT | NeurIPS 2020 | | |
+| Robust Federated Learning: The Case of Affine Distribution Shifts | MIT | NeurIPS 2020 | Privacy, Robustness | |
+| An Efficient Framework for Clustered Federated Learning | UCB | NeurIPS 2020 | heterogeneous data (non-I.I.D) | |
+| Distributionally Robust Federated Averaging | PSU | NeurIPS 2020 | Privacy, Robustness | |
+| Group Knowledge Transfer: Federated Learning of Large CNNs at the Edge | USC | NeurIPS 2020 | Efficient Training of Large DNN at Edge | |
+| A Scalable Approach for Privacy-Preserving Collaborative Machine Learning | USC | NeurIPS 2020 | Scalability | |
+| Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimization | CMU | NeurIPS 2020 | local update step heterogeneity | |
+| Attack of the Tails: Yes, You Really Can Backdoor Federated Learning | Wiscosin| NeurIPS 2020 | Privacy, Robustness | |
+| Federated Accelerated Stochastic Gradient Descent | Stanford | NeurIPS 2020 | Acceleration | |
+| Inverting Gradients - How easy is it to break privacy in federated learning? | University of Siegen | NeurIPS 2020 | Privacy, Robustness | |
+| Ensemble Distillation for Robust Model Fusion in Federated Learning | EPFL | NeurIPS 2020 | Privacy, Robustness | |
+| Optimal Topology Design for Cross-Silo Federated Learning | Inria | NeurIPS 2020 | Topology Optimization | |
+| Distributed Training with Heterogeneous Data: Bridging Median- and Mean-Based Algorithms | University of Minnesota | NeurIPS 2020 | | |
+| Distributed Distillation for On-Device Learning | Stanford | NeurIPS 2020 | | |
+| Byzantine Resilient Distributed Multi-Task Learning | Vanderbilt University | NeurIPS 2020 | | |
+| Distributed Newton Can Communicate Less and Resist Byzantine Workers | UCB | NeurIPS 2020 | | |
+| Minibatch vs Local SGD for Heterogeneous Distributed Learning | TTIC | NeurIPS 2020 | | |
+| Election Coding for Distributed Learning: Protecting SignSGD against Byzantine Attacks | | NeurIPS 2020 | | |
+
+(according to https://neurips.cc/Conferences/2020/AcceptedPapersInitial)
+
+Note: most of the accepted publications are preparing the camera ready revision, thus we are not sure the detail of their proposed methods
+
+
+## Research Areas
+#### Statistical Challenges: data distribution heterogeneity and label deficiency (159)
+ - [Distributed Optimization](#Distributed-optimization (68))
+ - [Non-IID and Model Personalization](#Non-IID-and-Model-Personalization (53))
+ - [Semi-Supervised Learning](#Semi-Supervised-Learning (3))
+ - [Vertical Federated Learning](#Vertical-Federated-Learning (8))
+ - [Decentralized FL](#Decentralized-FL (7))
+ - [Hierarchical FL](#Hierarchical-FL (8))
+ - [Neural Architecture Search](#Neural-Architecture-Search (4))
+ - [Transfer Learning](#Transfer-Learning (11))
+ - [Continual Learning](#continual-learning (1))
+ - [Domain Adaptation](#Domain-Adaptation)
+ - [Reinforcement Learning](#Reinforcement-Learning)
+ - [Bayesian Learning ](#Bayesian-Learning )
+ - [Causal Learning](#Causal-Learning )
+
+
+#### Trustworthiness: security, privacy, fairness, incentive mechanism, etc. (88)
+ - [Adversarial-Attack-and-Defense](#Adversarial-Attack-and-Defense)
+ - [Privacy](#Privacy (36))
+ - [Fairness](#Fairness (4))
+ - [Interpretability](#Interpretability)
+ - [Incentive Mechanism](#Incentive-Mechanism (5))
+
+#### System Challenges: communication and computational resource constrained, software and hardware heterogeneity, and FL system (141)
+ - [Communication-Efficiency](#Communication-Efficiency (29))
+ - [Straggler Problem](#straggler-problem (4))
+ - [Computation Efficiency](#Computation-Efficiency (14))
+ - [Wireless Communication and Cloud Computing](#Wireless-Communication-and-Cloud-Computing (74))
+ - [FL System Design](#FL-System-Design (20))
+
+#### Models and Applications (104)
+ - [Models](#Models (22))
+ - [Natural language Processing](#Natural-language-Processing (15))
+ - [Computer Vision](#Computer-Vision (3))
+ - [Health Care](#Health-Care (27))
+ - [Transportation](#Transportation (14))
+ - [Recommendation System](#Recommendation-System (8))
+ - [Speech](#Speech (1))
+ - [Finance](#Finance (2))
+ - [Smart City](#Smart-City (2))
+ - [Robotics](#Robotics (2))
+ - [Networking](#Networking (1))
+ - [Blockchain](#Blockchain (2))
+ - [Other](#Other (5))
+
+#### Benchmark, Dataset and Survey (27)
+ - [Benchmark and Dataset](#Benchmark-and-Dataset) (7)
+ - [Survey](#Survey) (20)
+
+-------------------
+
+# Statistical Challenges: distribution heterogeneity and label deficiency
+
+## Distributed optimization
+Userful Federated Optimizer Baselines:
+
+FedAvg:
+[Communication-Efficient Learning of Deep Networks from Decentralized Data. 2016-02. AISTAT 2017.](https://arxiv.org/pdf/1602.05629.pdf)
+
+FedOpt:
+[Adaptive Federated Optimization. ICLR 2021 (Under Review). 2020-02-29](https://arxiv.org/pdf/2003.00295.pdf)
+
+FedNov:
+[Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimization. NeurIPS 2020](https://arxiv.org/abs/2007.07481)
+
+-------------------------
+
+[Federated Optimization: Distributed Optimization Beyond the Datacenter. NIPS 2016 workshop.](https://arxiv.org/pdf/1511.03575.pdf)
+
+[Federated Optimization: Distributed Machine Learning for On-Device Intelligence](https://arxiv.org/pdf/1610.02527.pdf)
+
+[Stochastic, Distributed and Federated Optimization for Machine Learning. FL PhD Thesis. By Jakub](https://arxiv.org/pdf/1707.01155.pdf)
+
+[Collaborative Deep Learning in Fixed Topology Networks](https://arxiv.org/pdf/1706.07880.pdf)
+
+[Federated Multi-Task Learning](https://arxiv.org/pdf/1705.10467.pdf)
+
+[LAG: Lazily Aggregated Gradient for Communication-Efficient Distributed Learning](https://arxiv.org/abs/1805.09965)
+
+[Local Stochastic Approximation: A Unified View of Federated Learning and Distributed Multi-Task Reinforcement Learning Algorithms](https://arxiv.org/pdf/2006.13460.pdf)
+
+[Proxy Experience Replay: Federated Distillation for Distributed Reinforcement Learning](https://arxiv.org/pdf/2005.06105.pdf)
+
+[Exact Support Recovery in Federated Regression with One-shot Communication](https://arxiv.org/pdf/2006.12583.pdf)
+
+[DEED: A General Quantization Scheme for Communication Efficiency in Bits](https://arxiv.org/pdf/2006.11401.pdf)
+Researcher: Ruoyu Sun, UIUC
+
+[Robust Federated Learning: The Case of Affine Distribution Shifts](https://arxiv.org/pdf/2006.08907.pdf)
+
+[Personalized Federated Learning with Moreau Envelopes](https://arxiv.org/pdf/2006.08848.pdf)
+
+[Towards Flexible Device Participation in Federated Learning for Non-IID Data](https://arxiv.org/pdf/2006.06954.pdf)
+Keywords: inactive or return incomplete updates in non-IID dataset
+
+[A Primal-Dual SGD Algorithm for Distributed Nonconvex Optimization](https://arxiv.org/pdf/2006.03474.pdf)
+
+[FedPD: A Federated Learning Framework with Optimal Rates and Adaptivity to Non-IID Data](https://arxiv.org/pdf/2005.11418.pdf)
+Researcher: Wotao Yin, UCLA
+
+[FedSplit: An algorithmic framework for fast federated optimization](https://arxiv.org/pdf/2005.05238.pdf)
+
+[Distributed Stochastic Non-Convex Optimization: Momentum-Based Variance Reduction](https://arxiv.org/pdf/2005.00224.pdf)
+
+[On the Outsized Importance of Learning Rates in Local Update Methods](https://arxiv.org/pdf/2007.00878.pdf)
+Highlight: local model learning rate optimization + automation
+Researcher: Jakub
+
+[Federated Learning with Compression: Unified Analysis and Sharp Guarantees](https://arxiv.org/pdf/2007.01154.pdf)
+Highlight: non-IID, gradient compression + local SGD
+Researcher: Mehrdad Mahdavi, Jin Rong’s PhD Student http://www.cse.psu.edu/~mzm616/
+
+[From Local SGD to Local Fixed-Point Methods for Federated Learning](https://arxiv.org/pdf/2004.01442.pdf)
+
+[Federated Residual Learning. 2020-03](https://arxiv.org/pdf/2003.12880.pdf)
+
+
+[Acceleration for Compressed Gradient Descent in Distributed and Federated Optimization. ICML 2020.](https://arxiv.org/pdf/2002.11364.pdf)
+
+[LASG: Lazily Aggregated Stochastic Gradients for Communication-Efficient Distributed Learning](https://arxiv.org/pdf/2002.11360.pdf)
+
+[Uncertainty Principle for Communication Compression in Distributed and Federated Learning and the Search for an Optimal Compressor](https://arxiv.org/pdf/2002.08958.pdf)
+
+[Dynamic Federated Learning](https://arxiv.org/pdf/2002.08782.pdf)
+
+[Distributed Optimization over Block-Cyclic Data](https://arxiv.org/pdf/2002.07454.pdf)
+
+[Distributed Non-Convex Optimization with Sublinear Speedup under Intermittent Client Availability](https://arxiv.org/pdf/2002.07399.pdf)
+
+[Federated Learning with Matched Averaging](https://arxiv.org/pdf/2002.06440.pdf)
+
+[Federated Learning of a Mixture of Global and Local Models](https://arxiv.org/pdf/2002.05516.pdf)
+
+[Faster On-Device Training Using New Federated Momentum Algorithm](https://arxiv.org/pdf/2002.02090.pdf)
+
+[FedDANE: A Federated Newton-Type Method](https://arxiv.org/pdf/2001.01920.pdf)
+
+[Distributed Fixed Point Methods with Compressed Iterates](https://arxiv.org/pdf/1912.09925.pdf)
+
+[Primal-dual methods for large-scale and distributed convex optimization and data analytics](https://arxiv.org/pdf/1912.08546.pdf)
+
+[Parallel Restarted SPIDER - Communication Efficient Distributed Nonconvex Optimization with Optimal Computation Complexity](https://arxiv.org/pdf/1912.06036.pdf)
+
+[Representation of Federated Learning via Worst-Case Robust Optimization Theory](https://arxiv.org/pdf/1912.05571.pdf)
+
+[On the Convergence of Local Descent Methods in Federated Learning](https://arxiv.org/pdf/1910.14425.pdf)
+
+[SCAFFOLD: Stochastic Controlled Averaging for Federated Learning](https://arxiv.org/pdf/1910.06378.pdf)
+
+[Central Server Free Federated Learning over Single-sided Trust Social Networks](https://arxiv.org/pdf/1910.04956.pdf)
+
+[Accelerating Federated Learning via Momentum Gradient Descent](https://arxiv.org/pdf/1910.03197.pdf)
+
+[Communication-Efficient Distributed Optimization in Networks with Gradient Tracking and Variance Reduction](https://arxiv.org/pdf/1909.05844.pdf)
+
+[Gradient Descent with Compressed Iterates](https://arxiv.org/pdf/1909.04716.pdf)
+
+[First Analysis of Local GD on Heterogeneous Data](https://arxiv.org/pdf/1909.04715.pdf)
+
+[(*) On the Convergence of FedAvg on Non-IID Data. ICLR 2020.](https://arxiv.org/pdf/1907.02189.pdf)
+
+[Robust Federated Learning in a Heterogeneous Environment](https://arxiv.org/pdf/1906.06629.pdf)
+
+[Scalable and Differentially Private Distributed Aggregation in the Shuffled Model](https://arxiv.org/pdf/1906.08320.pdf)
+
+[Variational Federated Multi-Task Learning](https://arxiv.org/pdf/1906.06268.pdf)
+
+[Bayesian Nonparametric Federated Learning of Neural Networks. ICLR 2019.](https://arxiv.org/pdf/1905.12022.pdf)
+
+[Differentially Private Learning with Adaptive Clipping](https://arxiv.org/pdf/1905.03871.pdf)
+
+[Semi-Cyclic Stochastic Gradient Descent](https://arxiv.org/pdf/1904.10120.pdf)
+
+[Asynchronous Federated Optimization](https://arxiv.org/pdf/1903.03934.pdf)
+
+[Agnostic Federated Learning](https://arxiv.org/pdf/1902.00146.pdf)
+
+[Federated Optimization in Heterogeneous Networks](https://arxiv.org/pdf/1812.06127.pdf)
+
+[Partitioned Variational Inference: A unified framework encompassing federated and continual learning](https://arxiv.org/pdf/1811.11206.pdf)
+
+[Learning Rate Adaptation for Federated and Differentially Private Learning](https://arxiv.org/pdf/1809.03832.pdf)
+
+[Communication-Efficient Robust Federated Learning Over Heterogeneous Datasets](https://arxiv.org/pdf/2006.09992.pdf)
+
+[An Efficient Framework for Clustered Federated Learning](https://arxiv.org/pdf/2006.04088.pdf)
+
+[Adaptive Federated Learning in Resource Constrained Edge Computing Systems](https://arxiv.org/pdf/1804.05271.pdf)
+Citation: 146
+
+[Adaptive Federated Optimization](http://arxiv.org/pdf/2003.00295.pdf)
+
+[Local SGD converges fast and communicates little](https://arxiv.org/pdf/1805.09767.pdf)
+
+[Don’t Use Large Mini-Batches, Use Local SGD](https://arxiv.org/pdf/1808.07217.pdf)
+
+[Overlap Local-SGD: An Algorithmic Approach to Hide Communication Delays in Distributed SGD](https://arxiv.org/pdf/2002.09539.pdf)
+
+[Local SGD With a Communication Overhead Depending Only on the Number of Workers](https://arxiv.org/pdf/2006.02582.pdf)
+
+[Federated Accelerated Stochastic Gradient Descent ](https://arxiv.org/pdf/2006.08950.pdf)
+
+[Tighter Theory for Local SGD on Identical and Heterogeneous Data](https://arxiv.org/pdf/1909.04746.pdf)
+
+[STL-SGD: Speeding Up Local SGD with Stagewise Communication Period](https://arxiv.org/pdf/2006.06377.pdf)
+
+[Cooperative SGD: A unified Framework for the Design and Analysis of Communication-Efficient SGD Algorithms](https://arxiv.org/pdf/1808.07576.pdf)
+
+[Don't Use Large Mini-Batches, Use Local SGD](https://arxiv.org/pdf/1808.07217.pdf)
+
+[Understanding Unintended Memorization in Federated Learning](http://arxiv.org/pdf/2006.07490.pdf)
+
+## Non-IID and Model Personalization
+[The Non-IID Data Quagmire of Decentralized Machine Learning. 2019-10](https://arxiv.org/pdf/1910.00189.pdf)
+
+[Federated Learning with Non-IID Data](https://arxiv.org/pdf/1806.00582.pdf)
+
+[FedCD: Improving Performance in non-IID Federated Learning. 2020](https://arxiv.org/pdf/2006.09637.pdf)
+
+[Life Long Learning: FedFMC: Sequential Efficient Federated Learning on Non-iid Data. 2020](https://arxiv.org/pdf/2006.10937.pdf)
+
+[Robust Federated Learning: The Case of Affine Distribution Shifts. 2020](https://arxiv.org/pdf/2006.08907.pdf)
+
+[Personalized Federated Learning with Moreau Envelopes. 2020](https://arxiv.org/pdf/2006.08848.pdf)
+
+
+[Personalized Federated Learning using Hypernetworks. 2021](https://arxiv.org/pdf/2103.04628.pdf)
+
+[Ensemble Distillation for Robust Model Fusion in Federated Learning. 2020](https://arxiv.org/pdf/2006.07242.pdf)
+Researcher: Tao Lin, ZJU, EPFL https://tlin-tao-lin.github.io/index.html
+
+[Proxy Experience Replay: Federated Distillation for Distributed Reinforcement Learning. 2020](https://arxiv.org/pdf/2005.06105.pdf)
+
+[Towards Flexible Device Participation in Federated Learning for Non-IID Data. 2020](https://arxiv.org/pdf/2006.06954.pdf)
+Keywords: inactive or return incomplete updates in non-IID dataset
+
+[XOR Mixup: Privacy-Preserving Data Augmentation for One-Shot Federated Learning. 2020](https://arxiv.org/pdf/2006.05148.pdf)
+
+[NeurIPS 2020 submission: An Efficient Framework for Clustered Federated Learning. 2020](https://arxiv.org/pdf/2006.04088.pdf)
+Researcher: AVISHEK GHOSH, UCB, PhD
+
+[Continual Local Training for Better Initialization of Federated Models. 2020](https://arxiv.org/pdf/2005.12657.pdf)
+
+[FedPD: A Federated Learning Framework with Optimal Rates and Adaptivity to Non-IID Data. 2020](https://arxiv.org/pdf/2005.11418.pdf)
+Researcher: Wotao Yin, UCLA
+
+[Global Multiclass Classification from Heterogeneous Local Models. 2020](https://arxiv.org/pdf/2005.10848.pdf)
+Researcher: Stanford https://stanford.edu/~pilanci/
+
+[Multi-Center Federated Learning. 2020](https://arxiv.org/pdf/2005.01026.pdf)
+
+[Federated learning with hierarchical clustering of local updates to improve training on non-IID data. 2020](https://arxiv.org/pdf/2004.11791.pdf)
+
+[Federated Learning with Only Positive Labels. 2020](https://arxiv.org/pdf/2004.10342.pdf)
+Researcher: Felix Xinnan Yu, Google New York
+Keywords: positive labels
+Limited Labels
+
+[Federated Semi-Supervised Learning with Inter-Client Consistency. 2020](https://arxiv.org/pdf/2006.12097.pdf)
+
+[(*) FedMAX: Mitigating Activation Divergence for Accurate and Communication-Efficient Federated Learning. CMU ECE. 2020-04-07](https://arxiv.org/pdf/2004.03657.pdf)
+
+[(*) Adaptive Personalized Federated Learning](https://arxiv.org/pdf/2003.13461.pdf)
+
+[Semi-Federated Learning](https://arxiv.org/pdf/2003.12795.pdf)
+
+[Survey of Personalization Techniques for Federated Learning. 2020-03-19](https://arxiv.org/pdf/2003.08673.pdf)
+
+[Device Heterogeneity in Federated Learning: A Superquantile Approach. 2020-02](https://arxiv.org/pdf/2002.11223.pdf)
+
+[Personalized Federated Learning for Intelligent IoT Applications: A Cloud-Edge based Framework](https://arxiv.org/pdf/2002.10671.pdf)
+
+[Three Approaches for Personalization with Applications to Federated Learning](https://arxiv.org/pdf/2002.10619.pdf)
+
+[Personalized Federated Learning: A Meta-Learning Approach](https://arxiv.org/pdf/2002.07948.pdf)
+
+[Towards Federated Learning: Robustness Analytics to Data Heterogeneity](https://arxiv.org/pdf/2002.05038.pdf)
+Highlight: non-IID + adversarial attacks
+
+[Salvaging Federated Learning by Local Adaptation](https://arxiv.org/pdf/2002.04758.pdf)
+Highlight: an experimental paper that evaluate FL can help to improve the local accuracy
+
+[FOCUS: Dealing with Label Quality Disparity in Federated Learning. 2020-01](https://arxiv.org/pdf/2001.11359.pdf)
+
+[Overcoming Noisy and Irrelevant Data in Federated Learning. ICPR 2020.](https://arxiv.org/pdf/2001.08300.pdf)
+
+[Real-Time Edge Intelligence in the Making: A Collaborative Learning Framework via Federated Meta-Learning. 2020-01](https://arxiv.org/pdf/2001.03229.pdf)
+
+[(*) Think Locally, Act Globally: Federated Learning with Local and Global Representations. NeurIPS 2019 Workshop on Federated Learning distinguished student paper award](https://arxiv.org/pdf/2001.01523.pdf)
+
+[Federated Learning with Personalization Layers](https://arxiv.org/pdf/1912.00818.pdf)
+
+[Federated Adversarial Domain Adaptation](https://arxiv.org/pdf/1911.02054.pdf)
+
+[Federated Evaluation of On-device Personalization](https://arxiv.org/pdf/1910.10252.pdf)
+
+[Federated Learning with Unbiased Gradient Aggregation and Controllable Meta Updating](https://arxiv.org/pdf/1910.08234.pdf)
+
+[Overcoming Forgetting in Federated Learning on Non-IID Data](https://arxiv.org/pdf/1910.07796.pdf)
+
+[Clustered Federated Learning: Model-Agnostic Distributed Multi-Task Optimization under Privacy Constraints](https://arxiv.org/pdf/1910.01991.pdf)
+
+[Robust and Communication-Efficient Federated Learning From Non-i.i.d. Data](https://arxiv.org/pdf/1903.02891.pdf)
+
+[Improving Federated Learning Personalization via Model Agnostic Meta Learning](https://arxiv.org/pdf/1909.12488.pdf)
+
+[Measure Contribution of Participants in Federated Learning](https://arxiv.org/pdf/1909.08525.pdf)
+
+[(*) Measuring the Effects of Non-Identical Data Distribution for Federated Visual Classification](https://arxiv.org/pdf/1909.06335.pdf)
+
+[Multi-hop Federated Private Data Augmentation with Sample Compression](https://arxiv.org/pdf/1907.06426.pdf)
+
+[Astraea: Self-balancing Federated Learning for Improving Classification Accuracy of Mobile Deep Learning Applications](https://arxiv.org/pdf/1907.01132.pdf)
+
+[Distributed Training with Heterogeneous Data: Bridging Median- and Mean-Based Algorithms](https://arxiv.org/pdf/1906.01736.pdf)
+
+[Hybrid-FL for Wireless Networks: Cooperative Learning Mechanism Using Non-IID Data](https://arxiv.org/pdf/1905.07210.pdf)
+
+[Robust and Communication-Efficient Federated Learning from Non-IID Data](https://arxiv.org/pdf/1903.02891.pdf)
+
+[High Dimensional Restrictive Federated Model Selection with multi-objective Bayesian Optimization over shifted distributions](https://arxiv.org/pdf/1902.08999.pdf)
+
+[Client Selection for Federated Learning with Heterogeneous Resources in Mobile Edge](https://arxiv.org/pdf/1804.08333.pdf)
+
+[Federated Meta-Learning with Fast Convergence and Efficient Communication](https://arxiv.org/pdf/1802.07876.pdf)
+
+[Robust Federated Learning Through Representation Matching and Adaptive Hyper-parameters](https://arxiv.org/pdf/1912.13075.pdf)
+
+[Towards Efficient Scheduling of Federated Mobile Devices under Computational and Statistical Heterogeneity](https://arxiv.org/pdf/2005.12326.pdf)
+
+[Client Adaptation improves Federated Learning with Simulated Non-IID Clients](https://arxiv.org/pdf/2007.04806.pdf)
+
+[Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimization](https://arxiv.org/pdf/2007.07481.pdf)
+
+[Personalized Federated Learning by Structured and Unstructured Pruning under Data Heterogeneity. ICDCS 2021.](https://arxiv.org/abs/2105.00562)
+
+
+## Vertical Federated Learning
+[SecureBoost: A Lossless Federated Learning Framework](https://arxiv.org/pdf/1901.08755.pdf)
+
+[Parallel Distributed Logistic Regression for Vertical Federated Learning without Third-Party Coordinator](https://arxiv.org/pdf/1911.09824.pdf)
+
+[A Quasi-Newton Method Based Vertical Federated Learning Framework for Logistic Regression](https://arxiv.org/pdf/1912.00513.pdf)
+
+[Private federated learning on vertically partitioned data via entity resolution and additively homomorphic encryption](https://arxiv.org/pdf/1711.10677.pdf)
+
+[Entity Resolution and Federated Learning get a Federated Resolution.](https://arxiv.org/pdf/1803.04035.pdf)
+
+[Multi-Participant Multi-Class Vertical Federated Learning](https://arxiv.org/pdf/2001.11154.pdf)
+
+[A Communication-Efficient Collaborative Learning Framework for Distributed Features](https://arxiv.org/pdf/1912.11187.pdf)
+
+[Asymmetrical Vertical Federated Learning](https://arxiv.org/pdf/2004.07427.pdf)
+Researcher: Tencent Cloud, Libin Wang
+
+[VAFL: a Method of Vertical Asynchronous Federated Learning, ICML workshop on FL, 2020](https://arxiv.org/abs/2007.06081)
+
+
+## Decentralized FL
+[Central Server Free Federated Learning over Single-sided Trust Social Networks](https://arxiv.org/pdf/1910.04956.pdf)
+
+[Can Decentralized Algorithms Outperform Centralized Algorithms? A Case Study for Decentralized Parallel Stochastic Gradient Descent](https://arxiv.org/pdf/1705.09056.pdf)
+
+[Multi-consensus Decentralized Accelerated Gradient Descent](https://arxiv.org/pdf/2005.00797.pdf)
+
+[Decentralized Bayesian Learning over Graphs. 2019-05](https://arxiv.org/pdf/1905.10466.pdf)
+
+[BrainTorrent: A Peer-to-Peer Environment for Decentralized Federated Learning](https://arxiv.org/pdf/1905.06731.pdf)
+
+[Biscotti: A Ledger for Private and Secure Peer-to-Peer Machine Learning](https://arxiv.org/pdf/1811.09904.pdf)
+
+[Matcha: Speeding Up Decentralized SGD via Matching Decomposition Sampling](https://arxiv.org/pdf/1905.09435.pdf)
+
+
+## Hierarchical FL
+[Client-Edge-Cloud Hierarchical Federated Learning](https://arxiv.org/pdf/1905.06641.pdf)
+
+[(FL startup: Tongdun, HangZhou, China) Knowledge Federation: A Unified and Hierarchical Privacy-Preserving AI Framework. 2020-02](https://arxiv.org/pdf/2002.01647.pdf)
+
+[HFEL: Joint Edge Association and Resource Allocation for Cost-Efficient Hierarchical Federated Edge Learning](https://arxiv.org/pdf/2002.11343.pdf)
+
+[Hierarchical Federated Learning Across Heterogeneous Cellular Networks](https://arxiv.org/pdf/1909.02362.pdf)
+
+[Enhancing Privacy via Hierarchical Federated Learning](https://arxiv.org/pdf/2004.11361.pdf)
+
+[Federated learning with hierarchical clustering of local updates to improve training on non-IID data. 2020](https://arxiv.org/pdf/2004.11791.pdf)
+
+[Federated Hierarchical Hybrid Networks for Clickbait Detection](https://arxiv.org/pdf/1906.00638.pdf)
+
+[Matcha: Speeding Up Decentralized SGD via Matching Decomposition Sampling](https://arxiv.org/pdf/1905.09435.pdf) (in above section as well)
+
+## Neural Architecture Search
+[FedNAS: Federated Deep Learning via Neural Architecture Search. CVPR 2020. 2020-04-18](https://arxiv.org/pdf/2004.08546.pdf
+
+[Real-time Federated Evolutionary Neural Architecture Search. 2020-03](https://arxiv.org/pdf/2003.02793.pdf)
+
+[Federated Neural Architecture Search. 2020-06-14](https://arxiv.org/pdf/2002.06352.pdf)
+
+[Differentially-private Federated Neural Architecture Search. 2020-06](https://arxiv.org/pdf/2006.10559.pdf)
+
+## Transfer Learning
+
+[Communication-Efficient On-Device Machine Learning: Federated Distillation and Augmentation under Non-IID Private Data](https://arxiv.org/pdf/1811.11479.pdf)
+
+[Secure Federated Transfer Learning. IEEE Intelligent Systems 2018.](https://arxiv.org/pdf/1812.03337.pdf)
+
+
+[FedMD: Heterogenous Federated Learning via Model Distillation](https://arxiv.org/pdf/1910.03581.pdf)
+
+[Secure and Efficient Federated Transfer Learning](https://arxiv.org/pdf/1910.13271.pdf)
+
+[Wireless Federated Distillation for Distributed Edge Learning with Heterogeneous Data](https://arxiv.org/pdf/1907.02745.pdf)
+
+
+[Decentralized Differentially Private Segmentation with PATE. 2020-04](https://arxiv.org/pdf/2004.06567.pdf) \
+Highlights: apply the ICLR 2017 paper "Semisupervised knowledge transfer for deep learning from private training data"
+
+[Proxy Experience Replay: Federated Distillation for Distributed Reinforcement Learning. 2020](https://arxiv.org/pdf/2005.06105.pdf)
+
+[(FL startup: Tongdun, HangZhou, China) Knowledge Federation: A Unified and Hierarchical Privacy-Preserving AI Framework. 2020-02](https://arxiv.org/pdf/2002.01647.pdf)
+
+[Cooperative Learning via Federated Distillation over Fading Channels](https://arxiv.org/pdf/2002.01337.pdf)
+
+
+[(*) Cronus: Robust and Heterogeneous Collaborative Learning with Black-Box Knowledge Transfer](https://arxiv.org/pdf/1912.11279.pdf)
+
+[Federated Reinforcement Distillation with Proxy Experience Memory](https://arxiv.org/pdf/1907.06536.pdf)
+
+## Continual Learning
+[Federated Continual Learning with Adaptive Parameter Communication. 2020-03](https://arxiv.org/pdf/2003.03196.pdf)
+
+## Semi-Supervised Learning
+[Federated Semi-Supervised Learning with Inter-Client Consistency. 2020](https://arxiv.org/pdf/2006.12097.pdf)
+
+[Semi-supervised knowledge transfer for deep learning from private training data. ICLR 2017](https://arxiv.org/pdf/1610.05755.pdf)
+
+[Scalable private learning with PATE. ICLR 2018. ](https://arxiv.org/pdf/1802.08908.pdf)
+
+
+## Domain Adaptation
+[Federated Adversarial Domain Adaptation. ICLR 2020.](https://arxiv.org/pdf/1911.02054.pdf)
+
+## Reinforcement Learning
+[Federated Deep Reinforcement Learning](https://arxiv.org/pdf/1901.08277.pdf)
+
+## Bayesian Learning
+[Differentially Private Federated Variational Inference. NeurIPS 2019 FL Workshop. 2019-11-24.](https://arxiv.org/pdf/1911.10563.pdf)
+
+## Causal Learning
+[Towards Causal Federated Learning For Enhanced Robustness and Privacy. ICLR 2021 DPML Workshop](https://arxiv.org/pdf/2104.06557.pdf)
+
+# Trustworthy AI: adversarial attack, privacy, fairness, incentive mechanism, etc.
+
+## Adversarial Attack and Defense
+[An Overview of Federated Deep Learning Privacy Attacks and Defensive Strategies. 2020-04-01](https://arxiv.org/pdf/2004.04676.pdf)
+Citation: 0
+
+[How To Backdoor Federated Learning. 2018-07-02. AISTATS 2020](https://arxiv.org/pdf/1807.00459.pdf)
+Citation: 128
+
+[Can You Really Backdoor Federated Learning?. NeruIPS 2019. 2019-11-18](https://arxiv.org/pdf/1911.07963.pdf)
+Highlight: by Google
+Citation: 9
+
+[DBA: Distributed Backdoor Attacks against Federated Learning. ICLR 2020.](https://openreview.net/pdf?id=rkgyS0VFvr)
+Citation: 66
+
+[CRFL: Certifiably Robust Federated Learning against Backdoor Attacks. ICML 2021.](https://arxiv.org/pdf/2106.08283.pdf)
+
+[Deep Models Under the GAN: Information Leakage from Collaborative Deep Learning. ACM CCS 2017. 2017-02-14](https://arxiv.org/pdf/1702.07464.pdf)
+Citation: 284
+
+[Byzantine-Robust Distributed Learning: Towards Optimal Statistical Rates](https://arxiv.org/pdf/1803.01498.pdf)
+Citation: 112
+
+[Deep Leakage from Gradients. NIPS 2019](https://papers.nips.cc/paper/9617-deep-leakage-from-gradients.pdf)
+Citation: 31
+
+[Comprehensive Privacy Analysis of Deep Learning: Passive and Active White-box Inference Attacks against Centralized and Federated Learning. 2018-12-03](https://arxiv.org/pdf/1812.00910.pdf)
+Citation: 46
+
+[Beyond Inferring Class Representatives: User-Level Privacy Leakage From Federated Learning. INFOCOM 2019](https://arxiv.org/pdf/1812.00535.pdf)
+Citation: 56
+Highlight: server-side attack
+
+[Analyzing Federated Learning through an Adversarial Lens. ICML 2019.](https://arxiv.org/pdf/1811.12470.pdf).
+Citation: 60
+Highlight: client attack
+
+[Mitigating Sybils in Federated Learning Poisoning. 2018-08-14. RAID 2020](https://arxiv.org/pdf/1808.04866.pdf)
+Citation: 41
+Highlight: defense
+
+[RSA: Byzantine-Robust Stochastic Aggregation Methods for Distributed Learning from Heterogeneous Datasets, AAAI 2019](https://arxiv.org/abs/1811.03761)
+Citation: 34
+
+[(*) A Framework for Evaluating Gradient Leakage Attacks in Federated Learning. 2020-04-22](https://arxiv.org/pdf/2004.10397.pdf)
+Researcher: Wenqi Wei, Ling Liu, GaTech
+
+[(*) Local Model Poisoning Attacks to Byzantine-Robust Federated Learning. 2019-11-26](https://arxiv.org/pdf/1911.11815.pdf)
+
+[NeurIPS 2020 Submission: Backdoor Attacks on Federated Meta-Learning](https://arxiv.org/pdf/2006.07026.pdf)
+Researcher: Chien-Lun Chen, USC
+
+[Towards Realistic Byzantine-Robust Federated Learning. 2020-04-10](https://arxiv.org/pdf/2004.04986.pdf)
+
+[Data Poisoning Attacks on Federated Machine Learning. 2020-04-19](https://arxiv.org/pdf/2004.10020.pdf)
+
+[Exploiting Defenses against GAN-Based Feature Inference Attacks in Federated Learning. 2020-04-27](https://arxiv.org/pdf/2004.12571.pdf)
+
+[Byzantine-Resilient High-Dimensional SGD with Local Iterations on Heterogeneous Data. 2020-06-22](https://arxiv.org/pdf/2006.13041.pdf)
+Researcher: Suhas Diggavi, UCLA (https://scholar.google.com/citations?hl=en&user=hjTzNuQAAAAJ&view_op=list_works&sortby=pubdate)
+
+[(*) NeurIPS 2020 submission: FedMGDA+: Federated Learning meets Multi-objective Optimization. 2020-06-20](https://arxiv.org/pdf/2006.11489.pdf)
+
+[(*) NeurIPS 2020 submission: Free-rider Attacks on Model Aggregation in Federated Learning. 2020-06-26](https://arxiv.org/pdf/2006.11901.pdf)
+
+[FDA3 : Federated Defense Against Adversarial Attacks for Cloud-Based IIoT Applications. 2020-06-28](https://arxiv.org/pdf/2006.15632.pdf)
+
+
+[Privacy-preserving Weighted Federated Learning within Oracle-Aided MPC Framework. 2020-05-17](https://arxiv.org/pdf/2003.07630.pdf)
+Citation: 0
+
+[BASGD: Buffered Asynchronous SGD for Byzantine Learning. 2020-03-02](https://arxiv.org/pdf/2003.00937.pdf)
+
+[Stochastic-Sign SGD for Federated Learning with Theoretical Guarantees. 2020-02-25](https://arxiv.org/pdf/2002.10940.pdf)
+Citation: 1
+
+[Learning to Detect Malicious Clients for Robust Federated Learning. 2020-02-01](https://arxiv.org/pdf/2002.00211.pdf)
+
+[Robust Aggregation for Federated Learning. 2019-12-31](https://arxiv.org/pdf/1912.13445.pdf)
+Citation: 9
+
+[Towards Deep Federated Defenses Against Malware in Cloud Ecosystems. 2019-12-27](https://arxiv.org/pdf/1912.12370.pdf)
+
+[Attack-Resistant Federated Learning with Residual-based Reweighting. 2019-12-23](https://arxiv.org/pdf/1912.11464.pdf)
+
+[Cronus: Robust and Heterogeneous Collaborative Learning with Black-Box Knowledge Transfer. 2019-12-24](https://arxiv.org/pdf/1912.11279.pdf)
+Citation: 1
+
+[Free-riders in Federated Learning: Attacks and Defenses. 2019-11-28](https://arxiv.org/pdf/1911.12560.pdf)
+
+[Robust Federated Learning with Noisy Communication. 2019-11-01](https://arxiv.org/pdf/1911.00251.pdf)
+Citation: 4
+
+[Abnormal Client Behavior Detection in Federated Learning. 2019-10-22](https://arxiv.org/pdf/1910.09933.pdf)
+Citation: 3
+
+[Eavesdrop the Composition Proportion of Training Labels in Federated Learning. 2019-10-14](https://arxiv.org/pdf/1910.06044.pdf)
+Citation: 0
+
+[Byzantine-Robust Federated Machine Learning through Adaptive Model Averaging. 2019-09-11](https://arxiv.org/pdf/1909.05125.pdf)
+
+[An End-to-End Encrypted Neural Network for Gradient Updates Transmission in Federated Learning. 2019-08-22](https://arxiv.org/pdf/1908.08340.pdf)
+
+[Secure Distributed On-Device Learning Networks With Byzantine Adversaries. 2019-06-03](https://arxiv.org/pdf/1906.00887.pdf)
+Citation: 3
+
+[Robust Federated Training via Collaborative Machine Teaching using Trusted Instances. 2019-05-03](https://arxiv.org/pdf/1905.02941.pdf)
+Citation: 2
+
+[Dancing in the Dark: Private Multi-Party Machine Learning in an Untrusted Setting. 2018-11-23](https://arxiv.org/pdf/1811.09712.pdf)
+Citation: 4
+
+[Inverting Gradients - How easy is it to break privacy in federated learning? 2020-03-31](https://arxiv.org/pdf/2003.14053.pdf)
+Citation: 3
+
+[Quantification of the Leakage in Federated Learning. 2019-10-12](https://arxiv.org/pdf/1910.05467.pdf)
+Citation: 1
+
+## Privacy
+[Practical Secure Aggregation for Federated Learning on User-Held Data. NIPS 2016 workshop](https://arxiv.org/pdf/1611.04482.pdf)
+Highlight: cryptology
+
+[Differentially Private Federated Learning: A Client Level Perspective. NIPS 2017 Workshop](https://arxiv.org/pdf/1712.07557.pdf)
+
+[Exploiting Unintended Feature Leakage in Collaborative Learning. S&P 2019. 2018-05-10](https://arxiv.org/pdf/1805.04049.pdf)
+Citation: 105
+
+[(x) Gradient-Leaks: Understanding and Controlling Deanonymization in Federated Learning. 2018-05](https://arxiv.org/pdf/1805.05838.pdf)
+
+[A Hybrid Approach to Privacy-Preserving Federated Learning. AISec 2019. 2018-12-07](https://arxiv.org/pdf/1812.03224.pdf)
+Citation: 35
+
+[A generic framework for privacy preserving deep learning. PPML 2018. 2018-11-09](https://arxiv.org/pdf/1811.04017.pdf)
+Citation: 36
+
+[Federated Generative Privacy. IJCAI 2019 FL workshop. 2019-10-08](https://arxiv.org/pdf/1910.08385.pdf)
+Citation: 4
+
+[Enhancing the Privacy of Federated Learning with Sketching. 2019-11-05](https://arxiv.org/pdf/1911.01812.pdf)
+Citaiton: 0
+
+[Federated Learning with Bayesian Differential Privacy. 2019-11-22](https://arxiv.org/pdf/1911.10071.pdf)
+Citation: 5
+
+HybridAlpha: An Efficient Approach for Privacy-Preserving Federated Learning. AISec 2019. 2019-12-12
+[https://aisec.cc/](https://arxiv.org/pdf/1912.05897.pdf)
+
+[Private Federated Learning with Domain Adaptation. NeurIPS 2019 FL workshop. 2019-12-13](https://arxiv.org/pdf/1912.06733.pdf)
+
+[iDLG: Improved Deep Leakage from Gradients. 2020-01-08](https://arxiv.org/pdf/2001.02610.pdf)
+Citation: 3
+
+[Anonymizing Data for Privacy-Preserving Federated Learning. 2020-02-21](https://arxiv.org/pdf/2002.09096.pdf)
+
+[Practical and Bilateral Privacy-preserving Federated Learning. 2020-02-23](https://arxiv.org/pdf/2002.09843.pdf)
+Citation: 0
+
+[Decentralized Policy-Based Private Analytics. 2020-03-14](https://arxiv.org/pdf/2003.06612.pdf)
+Citation: 0
+
+[FedSel: Federated SGD under Local Differential Privacy with Top-k Dimension Selection. DASFAA 2020. 2020-03-24](https://arxiv.org/pdf/2003.10637.pdf)
+Citation: 0
+
+[Learn to Forget: User-Level Memorization Elimination in Federated Learning. 2020-03-24](https://arxiv.org/pdf/2003.10933.pdf)
+
+[LDP-Fed: Federated Learning with Local Differential Privacy. EdgeSys 2020. 2020-04-01](https://arxiv.org/pdf/2006.03637.pdf)
+Researcher: Ling Liu, GaTech
+Citation: 1
+
+[PrivFL: Practical Privacy-preserving Federated Regressions on High-dimensional Data over Mobile Networks. 2020-04-05](https://arxiv.org/pdf/2004.02264.pdf)
+Citation: 0
+
+[Local Differential Privacy based Federated Learning for Internet of Things. 2020-04-09](https://arxiv.org/pdf/2004.08856.pdf)
+Citation: 0
+
+[Differentially Private AirComp Federated Learning with Power Adaptation Harnessing Receiver Noise. 2020-04.](https://arxiv.org/pdf/2004.06337.pdf)
+
+[Decentralized Differentially Private Segmentation with PATE. MICCAI 2020 Under Review. 2020-04](https://arxiv.org/pdf/2004.06567.pdf) \
+Highlights: apply the ICLR 2017 paper "Semisupervised knowledge transfer for deep learning from private training data"
+
+
+[Enhancing Privacy via Hierarchical Federated Learning. 2020-04-23](https://arxiv.org/pdf/2004.11361.pdf)
+
+[Privacy Preserving Distributed Machine Learning with Federated Learning. 2020-04-25](https://arxiv.org/pdf/2004.12108.pdf)
+Citation: 0
+
+[Exploring Private Federated Learning with Laplacian Smoothing. 2020-05-01](https://arxiv.org/pdf/2005.00218.pdf)
+Citation: 0
+
+[Information-Theoretic Bounds on the Generalization Error and Privacy Leakage in Federated Learning. 2020-05-05](https://arxiv.org/pdf/2005.02503.pdf)
+Citation: 0
+
+[Efficient Privacy Preserving Edge Computing Framework for Image Classification. 2020-05-10](https://arxiv.org/pdf/2005.04563.pdf)
+Citation: 0
+
+[A Distributed Trust Framework for Privacy-Preserving Machine Learning. 2020-06-03](https://arxiv.org/pdf/2006.02456.pdf)
+Citation: 0
+
+[Secure Byzantine-Robust Machine Learning. 2020-06-08](https://arxiv.org/pdf/2006.04747.pdf)
+
+[ARIANN: Low-Interaction Privacy-Preserving Deep Learning via Function Secret Sharing. 2020-06-08](https://arxiv.org/pdf/2006.04593.pdf)
+
+[Privacy For Free: Wireless Federated Learning Via Uncoded Transmission With Adaptive Power Control. 2020-06-09](https://arxiv.org/pdf/2006.05459.pdf)
+Citation: 0
+
+[(*) Distributed Differentially Private Averaging with Improved Utility and Robustness to Malicious Parties. 2020-06-12](https://arxiv.org/pdf/2006.07218.pdf)
+Citation: 0
+
+[GS-WGAN: A Gradient-Sanitized Approach for Learning Differentially Private Generators. 2020-06-15](https://arxiv.org/pdf/2006.08848.pdf)
+Citation: 0
+
+[Federated Learning with Differential Privacy:Algorithms and Performance Analysis](https://arxiv.org/pdf/1911.00222.pdf)
+Citation: 2
+
+## Fairness
+[Fair Resource Allocation in Federated Learning. ICLR 2020.](https://arxiv.org/pdf/1905.10497.pdf)
+
+[Hierarchically Fair Federated Learning](https://arxiv.org/pdf/2004.10386.pdf)
+
+[Towards Fair and Privacy-Preserving Federated Deep Models](https://arxiv.org/pdf/1906.01167.pdf)
+
+## Interpretability
+[Interpret Federated Learning with Shapley Values. ](https://arxiv.org/pdf/1905.04519.pdf)
+
+
+## Incentive Mechanism
+
+[Record and reward federated learning contributions with blockchain. IEEE CyberC 2019](https://mblocklab.com/RecordandReward.pdf)
+
+[FMore: An Incentive Scheme of Multi-dimensional Auction for Federated Learning in MEC. ICDCS 2020](https://arxiv.org/pdf/2002.09699.pdf)
+
+[Toward an Automated Auction Framework for Wireless Federated Learning Services Market](https://arxiv.org/pdf/1912.06370.pdf)
+
+[Federated Learning for Edge Networks: Resource Optimization and Incentive Mechanism](https://arxiv.org/pdf/1911.05642.pdf)
+
+[Motivating Workers in Federated Learning: A Stackelberg Game Perspective](https://arxiv.org/pdf/1908.03092.pdf)
+
+[Incentive Design for Efficient Federated Learning in Mobile Networks: A Contract Theory Approach](https://arxiv.org/pdf/1905.07479.pdf)
+
+[A Learning-based Incentive Mechanism forFederated Learning](https://www.u-aizu.ac.jp/~pengli/files/fl_incentive_iot.pdf)
+
+[A Crowdsourcing Framework for On-Device Federated Learning](https://arxiv.org/pdf/1911.01046.pdf)
+
+# System Challenges: communication and computational resource constrained, software and hardware heterogeneity, and FL wireless communication system
+
+## Communication Efficiency
+[Federated Learning: Strategies for Improving Communication Efficiency](https://arxiv.org/pdf/1610.05492.pdf)
+Highlights: optimization
+
+[Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training. ICLR 2018. 2017-12-05](https://arxiv.org/pdf/1712.01887.pdf)
+Highlights: gradient compression
+Citation: 298
+
+[NeurIPS 2020 submission: Artemis: tight convergence guarantees for bidirectional compression in Federated Learning. 2020-06-25](https://arxiv.org/pdf/2006.14591.pdf)
+Highlights: bidirectional gradient compression
+
+[Scheduling Policy and Power Allocation for Federated Learning in NOMA Based MEC. 2020-06-21](https://arxiv.org/pdf/2006.13044.pdf)
+
+[(x) Federated Mutual Learning. 2020-06-27](https://arxiv.org/pdf/2006.16765.pdf)
+Highlights: Duplicate to Deep Mutual Learning. CVPR 2018
+
+[A Better Alternative to Error Feedback for Communication-Efficient Distributed Learning. 2020-06-19](https://arxiv.org/pdf/2006.11077.pdf)
+Researcher: Peter Richtárik
+
+[Federated Learning With Quantized Global Model Updates. 2020-06-18](https://arxiv.org/pdf/2006.10672.pdf)
+Researcher: Mohammad Mohammadi Amiri, Princeton, Information Theory and Machine Learning
+Highlights: model compression
+
+[Federated Learning with Compression: Unified Analysis and Sharp Guarantees. 2020-07-02](https://arxiv.org/pdf/2007.01154.pdf)
+Highlight: non-IID, gradient compression + local SGD
+Researcher: Mehrdad Mahdavi, Jin Rong’s PhD http://www.cse.psu.edu/~mzm616/
+
+[Evaluating the Communication Efficiency in Federated Learning Algorithm. 2020-04-06](https://arxiv.org/pdf/2004.02738.pdf)
+
+[Dynamic Sampling and Selective Masking for Communication-Efficient Federated Learning. 2020-05-21](https://arxiv.org/pdf/2003.09603.pdf)
+
+[Ternary Compression for Communication-Efficient Federated Learning. 2020-05-07](https://arxiv.org/pdf/2003.03564.pdf)
+
+[Gradient Statistics Aware Power Control for Over-the-Air Federated Learning. 2020-05-04](https://arxiv.org/pdf/2003.02089.pdf)
+
+[Communication-Efficient Decentralized Learning with Sparsification and Adaptive Peer Selection. 2020-02-22](https://arxiv.org/pdf/2002.09692.pdf)
+
+[(*) RPN: A Residual Pooling Network for Efficient Federated Learning. ECAI 2020.](https://arxiv.org/pdf/2001.08600.pdf)
+
+[Intermittent Pulling with Local Compensation for Communication-Efficient Federated Learning. 2020-01-22](https://arxiv.org/pdf/2001.08277.pdf)
+
+[Hyper-Sphere Quantization: Communication-Efficient SGD for Federated Learning. 2019-11-12](https://arxiv.org/pdf/1911.04655.pdf)
+
+[L-FGADMM: Layer-Wise Federated Group ADMM for Communication Efficient Decentralized Deep Learning](https://arxiv.org/pdf/1911.03654.pdf)
+
+[Gradient Sparification for Asynchronous Distributed Training. 2019-10-24](https://arxiv.org/pdf/1910.10929.pdf)
+
+[High-Dimensional Stochastic Gradient Quantization for Communication-Efficient Edge Learning](https://arxiv.org/pdf/1910.03865.pdf)
+
+[SAFA: a Semi-Asynchronous Protocol for Fast Federated Learning with Low Overhead](https://arxiv.org/pdf/1910.01355.pdf)
+
+[Detailed comparison of communication efficiency of split learning and federated learning](https://arxiv.org/pdf/1909.09145.pdf)
+
+[Decentralized Federated Learning: A Segmented Gossip Approach](https://arxiv.org/pdf/1908.07782.pdf)
+
+[Communication-Efficient Federated Deep Learning with Asynchronous Model Update and Temporally Weighted Aggregation](https://arxiv.org/pdf/1903.07424.pdf)
+
+[One-Shot Federated Learning](https://arxiv.org/pdf/1902.11175.pdf)
+
+[Multi-objective Evolutionary Federated Learning](https://arxiv.org/pdf/1812.07478.pdf)
+
+[Expanding the Reach of Federated Learning by Reducing Client Resource Requirements](https://arxiv.org/pdf/1812.07210.pdf)
+
+[Partitioned Variational Inference: A unified framework encompassing federated and continual learning](https://arxiv.org/pdf/1811.11206.pdf)
+
+[FedOpt: Towards communication efficiency and privacy preservation in federated learning](https://res.mdpi.com/d_attachment/applsci/applsci-10-02864/article_deploy/applsci-10-02864.pdf)
+
+[A performance evaluation of federated learning algorithms](https://www.researchgate.net/profile/Gregor_Ulm/[publication/329106719_A_Performance_Evaluation_of_Federated_Learning_Algorithms]/(links/5c0fabcfa6fdcc494febf907/A-Performance-Evaluation-of-Federated-Learning-Algorithms.pdf))
+
+
+
+## Straggler Problem
+
+[Coded Federated Learning. Presented at the Wireless Edge Intelligence Workshop, IEEE GLOBECOM 2019](https://arxiv.org/pdf/2002.09574.pdf)
+
+[Turbo-Aggregate: Breaking the Quadratic Aggregation Barrier in Secure Federated Learning](https://arxiv.org/pdf/2002.04156.pdf)
+
+[Coded Federated Computing in Wireless Networks with Straggling Devices and Imperfect CSI](https://arxiv.org/pdf/1901.05239.pdf)
+
+[Information-Theoretic Perspective of Federated Learning](https://arxiv.org/pdf/1911.07652.pdf)
+
+
+## Computation Efficiency
+[NeurIPS 2020 Submission: Distributed Learning on Heterogeneous Resource-Constrained Devices](https://arxiv.org/pdf/2006.05403.pdf)
+
+[SplitFed: When Federated Learning Meets Split Learning](https://arxiv.org/pdf/2004.12088.pdf)
+
+[Lottery Hypothesis based Unsupervised Pre-training for Model Compression in Federated Learning](https://arxiv.org/pdf/2004.09817.pdf)
+
+[Secure Federated Learning in 5G Mobile Networks. 2020/04](https://arxiv.org/pdf/2004.06700.pdf)
+
+[ELFISH: Resource-Aware Federated Learning on Heterogeneous Edge Devices](https://arxiv.org/pdf/1912.01684.pdf)
+
+[Asynchronous Online Federated Learning for Edge Devices](https://arxiv.org/pdf/1911.02134.pdf)
+
+[(*) Secure Federated Submodel Learning](https://arxiv.org/pdf/1911.02254.pdf)
+
+[Federated Neuromorphic Learning of Spiking Neural Networks for Low-Power Edge Intelligence](https://arxiv.org/pdf/1910.09594.pdf)
+
+[Model Pruning Enables Efficient Federated Learning on Edge Devices](https://arxiv.org/pdf/1909.12326.pdf)
+
+[Towards Effective Device-Aware Federated Learning](https://arxiv.org/pdf/1908.07420.pdf)
+
+[Accelerating DNN Training in Wireless Federated Edge Learning System](https://arxiv.org/pdf/1905.09712.pdf)
+
+[Split learning for health: Distributed deep learning without sharing raw patient data](https://arxiv.org/pdf/1812.00564.pdf)
+
+[SmartPC: Hierarchical pace control in real-time federated learning system](https://www.ece.ucf.edu/~zsguo/pubs/conference_workshop/RTSS2019b.pdf)
+
+[DeCaf: Iterative collaborative processing over the edge](https://www.usenix.org/system/files/hotedge19-paper-kumar.pdf)
+
+## Wireless Communication and Cloud Computing
+Researcher:
+H. Vincent Poor
+https://ee.princeton.edu/people/h-vincent-poor
+
+Hao Ye
+https://scholar.google.ca/citations?user=ok7OWEAAAAAJ&hl=en
+
+Ye Li
+http://liye.ece.gatech.edu/
+
+[Mix2FLD: Downlink Federated Learning After Uplink Federated Distillation With Two-Way Mixup](https://arxiv.org/pdf/2006.09801.pdf)
+Researcher: Mehdi Bennis, Seong-Lyun Kim
+
+[Wireless Communications for Collaborative Federated Learning in the Internet of Things](https://arxiv.org/pdf/2006.02499.pdf)
+
+[Democratizing the Edge: A Pervasive Edge Computing Framework](https://arxiv.org/pdf/2007.00641.pdf)
+
+[UVeQFed: Universal Vector Quantization for Federated Learning](https://arxiv.org/pdf/2006.03262.pdf)
+
+[Federated Deep Learning Framework For Hybrid Beamforming in mm-Wave Massive MIMO](https://arxiv.org/pdf/2005.09969.pdf)
+
+[Efficient Federated Learning over Multiple Access Channel with Differential Privacy Constraints](https://arxiv.org/pdf/2005.07776.pdf)
+
+[A Secure Federated Learning Framework for 5G Networks](https://arxiv.org/pdf/2005.05752.pdf)
+
+[Federated Learning and Wireless Communications](https://arxiv.org/pdf/2005.05265.pdf)
+
+[Lightwave Power Transfer for Federated Learning-based Wireless Networks](https://arxiv.org/pdf/2005.03977.pdf)
+
+[Towards Ubiquitous AI in 6G with Federated Learning](https://arxiv.org/pdf/2004.13563.pdf)
+
+[Optimizing Over-the-Air Computation in IRS-Aided C-RAN Systems](https://arxiv.org/pdf/2004.09168.pdf)
+
+[Network-Aware Optimization of Distributed Learning for Fog Computing](https://arxiv.org/pdf/2004.08488.pdf)
+
+[On the Design of Communication Efficient Federated Learning over Wireless Networks](https://arxiv.org/pdf/2004.07351.pdf)
+
+[Federated Machine Learning for Intelligent IoT via Reconfigurable Intelligent Surface](https://arxiv.org/pdf/2004.05843.pdf)
+
+[Client Selection and Bandwidth Allocation in Wireless Federated Learning Networks: A Long-Term Perspective](https://arxiv.org/pdf/2004.04314.pdf)
+
+[Resource Management for Blockchain-enabled Federated Learning: A Deep Reinforcement Learning Approach](https://arxiv.org/pdf/2004.04104.pdf)
+
+[A Blockchain-based Decentralized Federated Learning Framework with Committee Consensus](https://arxiv.org/pdf/2004.00773.pdf)
+
+[Scheduling for Cellular Federated Edge Learning with Importance and Channel. 2020-04](https://arxiv.org/pdf/2004.00490.pdf)
+
+[Differentially Private Federated Learning for Resource-Constrained Internet of Things. 2020-03](https://arxiv.org/pdf/2003.12705.pdf)
+
+[Federated Learning for Task and Resource Allocation in Wireless High Altitude Balloon Networks. 2020-03](https://arxiv.org/pdf/2003.09375.pdf)
+
+[Gradient Estimation for Federated Learning over Massive MIMO Communication Systems](https://arxiv.org/pdf/2003.08059.pdf)
+
+[Adaptive Federated Learning With Gradient Compression in Uplink NOMA](https://arxiv.org/pdf/2003.01344.pdf)
+
+[Performance Analysis and Optimization in Privacy-Preserving Federated Learning](https://arxiv.org/pdf/2003.00229.pdf)
+
+[Energy-Efficient Federated Edge Learning with Joint Communication and Computation Design](https://arxiv.org/pdf/2003.00199.pdf)
+
+[Federated Over-the-Air Subspace Learning and Tracking from Incomplete Data](https://arxiv.org/pdf/2002.12873.pdf)
+
+[Decentralized Federated Learning via SGD over Wireless D2D Networks](https://arxiv.org/pdf/2002.12507.pdf)
+
+[HFEL: Joint Edge Association and Resource Allocation for Cost-Efficient Hierarchical Federated Edge Learning](https://arxiv.org/pdf/2002.11343.pdf)
+
+[Federated Learning in the Sky: Joint Power Allocation and Scheduling with UAV Swarms](https://arxiv.org/pdf/2002.08196.pdf)
+
+[Wireless Federated Learning with Local Differential Privacy](https://arxiv.org/pdf/2002.05151.pdf)
+
+[Cooperative Learning via Federated Distillation over Fading Channels](https://arxiv.org/pdf/2002.01337.pdf)
+
+[Federated Learning under Channel Uncertainty: Joint Client Scheduling and Resource Allocation. 2020-02](https://arxiv.org/pdf/2002.01337.pdf)
+
+[Learning from Peers at the Wireless Edge](https://arxiv.org/pdf/2001.11567.pdf)
+
+[Convergence of Update Aware Device Scheduling for Federated Learning at the Wireless Edge](https://arxiv.org/pdf/2001.10402.pdf)
+
+[Communication Efficient Federated Learning over Multiple Access Channels](https://arxiv.org/pdf/2001.08737.pdf)
+
+[Convergence Time Optimization for Federated Learning over Wireless Networks](https://arxiv.org/pdf/2001.07845.pdf)
+
+[One-Bit Over-the-Air Aggregation for Communication-Efficient Federated Edge Learning: Design and Convergence Analysis](https://arxiv.org/pdf/2001.05713.pdf)
+
+[Federated Learning with Cooperating Devices: A Consensus Approach for Massive IoT Networks. IEEE Internet of Things Journal. 2020](https://arxiv.org/pdf/1912.13163.pdf)
+
+[Asynchronous Federated Learning with Differential Privacy for Edge Intelligence](https://arxiv.org/pdf/1912.07902.pdf)
+
+[Federated learning with multichannel ALOHA](https://arxiv.org/pdf/1912.06273.pdf)
+
+[Federated Learning with Autotuned Communication-Efficient Secure Aggregation](https://arxiv.org/pdf/1912.00131.pdf)
+
+[Bandwidth Slicing to Boost Federated Learning in Edge Computing](https://arxiv.org/pdf/1911.07615.pdf)
+
+[Energy Efficient Federated Learning Over Wireless Communication Networks](https://arxiv.org/pdf/1911.02417.pdf)
+
+[Device Scheduling with Fast Convergence for Wireless Federated Learning](https://arxiv.org/pdf/1911.00856.pdf)
+
+[Energy-Aware Analog Aggregation for Federated Learning with Redundant Data](https://arxiv.org/pdf/1911.00188.pdf)
+
+[Age-Based Scheduling Policy for Federated Learning in Mobile Edge Networks](https://arxiv.org/pdf/1910.14648.pdf)
+
+[Federated Learning over Wireless Networks: Convergence Analysis and Resource Allocation](https://arxiv.org/pdf/1910.13067.pdf)
+
+[Federated Learning over Wireless Networks: Optimization Model Design and Analysis](http://networking.khu.ac.kr/layouts/net/publications/data/2019\)Federated%20Learning%20over%20Wireless%20Network.pdf)
+
+[Resource Allocation in Mobility-Aware Federated Learning Networks: A Deep Reinforcement Learning Approach](https://arxiv.org/pdf/1910.09172.pdf)
+
+[Reliable Federated Learning for Mobile Networks](https://arxiv.org/pdf/1910.06837.pdf)
+
+[FedPAQ: A Communication-Efficient Federated Learning Method with Periodic Averaging and Quantization](https://arxiv.org/pdf/1909.13014.pdf)
+
+[Active Federated Learning](https://arxiv.org/pdf/1909.12641.pdf)
+
+[Cell-Free Massive MIMO for Wireless Federated Learning](https://arxiv.org/pdf/1909.12567.pdf)
+
+[A Joint Learning and Communications Framework for Federated Learning over Wireless Networks](https://arxiv.org/pdf/1909.07972.pdf)
+
+[On Safeguarding Privacy and Security in the Framework of Federated Learning](https://arxiv.org/pdf/1909.06512.pdf)
+
+[On Safeguarding Privacy and Security in the Framework of Federated Learning](https://arxiv.org/pdf/1909.06512.pdf)
+
+[Hierarchical Federated Learning Across Heterogeneous Cellular Networks](https://arxiv.org/pdf/1909.02362.pdf)
+
+[Federated Learning for Wireless Communications: Motivation, Opportunities and Challenges](https://arxiv.org/pdf/1908.06847.pdf)
+
+[Scheduling Policies for Federated Learning in Wireless Networks](https://arxiv.org/pdf/1908.06287.pdf)
+
+[Federated Learning with Additional Mechanisms on Clients to Reduce Communication Costs](https://arxiv.org/pdf/1908.05891.pdf)
+
+[Federated Learning over Wireless Fading Channels](https://arxiv.org/pdf/1907.09769.pdf)
+
+[Energy-Efficient Radio Resource Allocation for Federated Edge Learning](https://arxiv.org/pdf/1907.06040.pdf)
+
+[Mobile Edge Computing, Blockchain and Reputation-based Crowdsourcing IoT Federated Learning: A Secure, Decentralized and Privacy-preserving System](https://arxiv.org/pdf/1906.10893.pdf)
+
+[Active Learning Solution on Distributed Edge Computing](https://arxiv.org/pdf/1906.10718.pdf)
+
+[Fast Uplink Grant for NOMA: a Federated Learning based Approach](https://arxiv.org/pdf/1905.04519.pdf)
+
+[Machine Learning at the Wireless Edge: Distributed Stochastic Gradient Descent Over-the-Air](https://arxiv.org/pdf/1901.00844.pdf)
+
+[Federated Learning via Over-the-Air Computation](https://arxiv.org/pdf/1812.11750.pdf)
+
+[Broadband Analog Aggregation for Low-Latency Federated Edge Learning](https://arxiv.org/pdf/1812.11494.pdf)
+
+[Federated Echo State Learning for Minimizing Breaks in Presence in Wireless Virtual Reality Networks](https://arxiv.org/pdf/1812.01202.pdf)
+
+[Joint Service Pricing and Cooperative Relay Communication for Federated Learning](https://arxiv.org/pdf/1811.12082.pdf)
+
+[In-Edge AI: Intelligentizing Mobile Edge Computing, Caching and Communication by Federated Learning](https://arxiv.org/pdf/1809.07857.pdf)
+
+[Asynchronous Task Allocation for Federated and Parallelized Mobile Edge Learning](https://arxiv.org/pdf/1905.01656.pdf)
+
+[CoLearn: enabling federated learning in MUD-compliant IoT edge networks](CoLearn: enabling federated learning in MUD-compliant IoT edge networks)
+
+## FL System Design
+[Towards Federated Learning at Scale: System Design](https://arxiv.org/pdf/1902.01046.pdf)
+
+[FedML: A Research Library and Benchmark for Federated Machine Learning](https://arxiv.org/pdf/2007.13518.pdf)
+
+[A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection](https://arxiv.org/pdf/1907.09693.pdf)
+
+[FLeet: Online Federated Learning via Staleness Awareness and Performance Prediction](https://arxiv.org/pdf/2006.07273.pdf)
+Researcher: Georgios Damaskinos, MLSys, https://people.epfl.ch/georgios.damaskinos?lang=en
+
+[Heterogeneity-Aware Federated Learning](https://arxiv.org/pdf/2006.06983.pdf)
+Researcher: Mengwei Xu, PKU
+
+Responsive Web User Interface to Recover Training Data from User Gradients in Federated Learning
+https://ldp-machine-learning.herokuapp.com/
+
+[Decentralised Learning from Independent Multi-Domain Labels for Person Re-Identification](https://arxiv.org/pdf/2006.04150.pdf)
+
+[[startup] Industrial Federated Learning -- Requirements and System Design](https://arxiv.org/pdf/2005.06850.pdf)
+
+[(startup) Federated Learning and Differential Privacy: Software tools analysis, the Sherpa.ai FL framework and methodological guidelines for preserving data privacy](https://arxiv.org/pdf/2007.00914.pdf)
+
+[(FL startup: Tongdun, HangZhou, China) Knowledge Federation: A Unified and Hierarchical Privacy-Preserving AI Framework. 2020-02](https://arxiv.org/pdf/2002.01647.pdf)
+
+[(*) TiFL: A Tier-based Federated Learning System. HPDC 2020 (High-Performance Parallel and Distributed Computing).](https://arxiv.org/pdf/2001.09249.pdf)
+
+[FMore: An Incentive Scheme of Multi-dimensional Auction for Federated Learning in MEC. ICDCS 2020 (2020 International Conference on Distributed Computing Systems)](https://arxiv.org/pdf/2002.09699.pdf)
+
+[Adaptive Gradient Sparsification for Efficient Federated Learning: An Online Learning Approach. ICDCS 2020 (2020 International Conference on Distributed Computing Systems)](https://arxiv.org/pdf/2001.04756.pdf)
+
+[Quantifying the Performance of Federated Transfer Learning](https://arxiv.org/pdf/1912.12795.pdf)
+
+[ELFISH: Resource-Aware Federated Learning on Heterogeneous Edge Devices](https://arxiv.org/pdf/1912.01684.pdf)
+
+[Privacy is What We Care About: Experimental Investigation of Federated Learning on Edge Devices](https://arxiv.org/pdf/1911.04559.pdf)
+
+[Substra: a framework for privacy-preserving, traceable and collaborative Machine Learning](https://arxiv.org/pdf/1910.11567.pdf)
+
+[BAFFLE : Blockchain Based Aggregator Free Federated Learning](https://arxiv.org/pdf/1909.07452.pdf)
+
+[Edge AIBench: Towards Comprehensive End-to-end Edge Computing Benchmarking](https://arxiv.org/pdf/1908.01924.pdf)
+
+[Functional Federated Learning in Erlang (ffl-erl)](https://arxiv.org/pdf/1808.08143.pdf)
+
+[HierTrain: Fast Hierarchical Edge AI Learning With Hybrid Parallelism in Mobile-Edge-Cloud Computing](https://arxiv.org/pdf/2003.09876.pdf)
+
+
+# Models and Applications
+
+## Models
+### Graph Neural Networks
+
+[Peer-to-peer federated learning on graphs](https://arxiv.org/pdf/1901.11173)
+
+[Towards Federated Graph Learning for Collaborative Financial Crimes Detection](https://arxiv.org/pdf/1909.12946)
+
+[A Graph Federated Architecture with Privacy Preserving Learning](https://arxiv.org/pdf/2104.13215)
+
+[Federated Myopic Community Detection with One-shot Communication](https://arxiv.org/pdf/2106.07255)
+
+[Federated Dynamic GNN with Secure Aggregation](https://arxiv.org/pdf/2009.07351)
+
+[Privacy-Preserving Graph Neural Network for Node Classification](https://arxiv.org/pdf/2005.11903)
+
+[ASFGNN: Automated Separated-Federated Graph Neural Network](https://arxiv.org/pdf/2011.03248)
+
+[GraphFL: A Federated Learning Framework for Semi-Supervised Node Classification on Graphs](https://arxiv.org/pdf/2012.04187)
+
+[FedGNN: Federated Graph Neural Network for Privacy-Preserving Recommendation](https://arxiv.org/pdf/2102.04925)
+
+[FedGraphNN: A Federated Learning System and Benchmark for Graph Neural Networks](https://arxiv.org/pdf/2104.07145)
+
+[FL-AGCNS: Federated Learning Framework for Automatic Graph Convolutional Network Search](https://arxiv.org/pdf/2104.04141)
+
+[Cluster-driven Graph Federated Learning over Multiple Domains](https://arxiv.org/pdf/2104.14628)
+
+[FedGL: Federated Graph Learning Framework with Global Self-Supervision](https://arxiv.org/pdf/2105.03170)
+
+[Federated Graph Learning -- A Position Paper](https://arxiv.org/pdf/2105.11099)
+
+[SpreadGNN: Serverless Multi-task Federated Learning for Graph Neural Networks](https://arxiv.org/pdf/2106.02743)
+
+[Cross-Node Federated Graph Neural Network for Spatio-Temporal Data Modeling](https://arxiv.org/pdf/2106.05223)
+
+[A Vertical Federated Learning Framework for Graph Convolutional Network](https://arxiv.org/pdf/2106.11593)
+
+[Federated Graph Classification over Non-IID Graphs](https://arxiv.org/pdf/2106.13423)
+
+[Subgraph Federated Learning with Missing Neighbor Generation](https://arxiv.org/pdf/2106.13430)
+
+### Federated Learning on Knowledge Graphs
+
+[FedE: Embedding Knowledge Graphs in Federated Setting](https://arxiv.org/pdf/2010.12882)
+
+[Improving Federated Relational Data Modeling via Basis Alignment and Weight Penalty](https://arxiv.org/pdf/2011.11369)
+
+[Federated Knowledge Graphs Embedding](https://arxiv.org/pdf/2105.07615)
+
+
+### Generative Models (GAN, Bayesian Generative Models, etc)
+
+[Discrete-Time Cox Models](https://arxiv.org/pdf/2006.08997.pdf)
+
+[Generative Models for Effective ML on Private, Decentralized Datasets. Google. ICLR 2020](https://arxiv.org/pdf/1911.06679.pdf)
+Citation: 8
+
+[MD-GAN: Multi-Discriminator Generative Adversarial Networks for Distributed Datasets. 2018-11-09](https://arxiv.org/pdf/1811.03850.pdf)
+
+[(GAN) Federated Generative Adversarial Learning. 2020-05-07](https://arxiv.org/pdf/2005.03793.pdf)
+Citation: 0
+
+[Differentially Private Data Generative Models](https://arxiv.org/pdf/1812.02274.pdf)
+
+[GRAFFL: Gradient-free Federated Learning of a Bayesian Generative Model](https://arxiv.org/pdf/1910.08489.pdf)
+
+### VAE (Variational Autoencoder)
+
+[(VAE) An On-Device Federated Learning Approach for Cooperative Anomaly Detection](https://arxiv.org/pdf/2002.12301.pdf)
+
+### MF (Matrix Factorization)
+
+[Secure Federated Matrix Factorization](https://arxiv.org/pdf/1906.05108.pdf)
+
+[(Clustering) Federated Clustering via Matrix Factorization Models: From Model Averaging to Gradient Sharing](https://arxiv.org/pdf/2002.04930.pdf)
+
+[Privacy Threats Against Federated Matrix Factorization](https://arxiv.org/pdf/2007.01587.pdf)
+
+### GBDT (Gradient Boosting Decision Trees)
+
+[Practical Federated Gradient Boosting Decision Trees. AAAI 2020.](https://arxiv.org/pdf/1911.04206.pdf)
+
+[Federated Extra-Trees with Privacy Preserving](https://arxiv.org/pdf/2002.07323.pdf)
+
+[SecureGBM: Secure Multi-Party Gradient Boosting](https://arxiv.org/pdf/1911.11997.pdf)
+
+[Federated Forest](https://arxiv.org/pdf/1905.10053.pdf)
+
+[The Tradeoff Between Privacy and Accuracy in Anomaly Detection Using Federated XGBoost](https://arxiv.org/pdf/1907.07157.pdf)
+
+### Other Model
+[Privacy Preserving QoE Modeling using Collaborative Learning](https://arxiv.org/pdf/1906.09248.pdf)
+
+
+[Distributed Dual Coordinate Ascent in General Tree Networks and Its Application in Federated Learning](https://arxiv.org/pdf/1703.04785.pdf)
+
+## Natural language Processing
+[Federated pretraining and fine tuning of BERT using clinical notes from multiple silos](https://arxiv.org/pdf/2002.08562.pdf)
+
+[Federated Learning for Mobile Keyboard Prediction](https://arxiv.org/pdf/1811.03604.pdf)
+
+[Federated Learning for Keyword Spotting](https://arxiv.org/pdf/1810.05512.pdf)
+
+[generative sequence models (e.g., language models)](https://arxiv.org/pdf/2006.07490.pdf)
+
+[Pretraining Federated Text Models for Next Word Prediction](https://arxiv.org/pdf/2005.04828.pdf)
+
+[FedNER: Privacy-preserving Medical Named Entity Recognition with Federated Learning. MSRA. 2020-03.](https://arxiv.org/pdf/2003.09288.pdf)
+
+[Federated Learning of N-gram Language Models. Google. ACL 2019.](https://www.aclweb.org/anthology/K19-1012.pdf)
+
+[Federated User Representation Learning](https://arxiv.org/pdf/1909.12535.pdf)
+
+[Two-stage Federated Phenotyping and Patient Representation Learning](https://arxiv.org/pdf/1908.05596.pdf)
+
+[Federated Learning for Emoji Prediction in a Mobile Keyboard](https://arxiv.org/pdf/1906.04329.pdf)
+
+[Federated AI lets a team imagine together: Federated Learning of GANs](https://arxiv.org/pdf/1906.03595.pdf)
+
+[Federated Learning Of Out-Of-Vocabulary Words](https://arxiv.org/pdf/1903.10635.pdf)
+
+[Learning Private Neural Language Modeling with Attentive Aggregation](https://arxiv.org/pdf/1812.07108.pdf)
+
+[Applied Federated Learning: Improving Google Keyboard Query Suggestions](https://arxiv.org/pdf/1812.02903.pdf)
+
+[Federated Learning for Ranking Browser History Suggestions](https://arxiv.org/pdf/1911.11807.pdf)
+
+## Computer Vision
+[Federated Face Anti-spoofing](https://arxiv.org/pdf/2005.14638.pdf)
+
+[(*) Federated Visual Classification with Real-World Data Distribution. MIT. ECCV 2020. 2020-03](https://arxiv.org/pdf/2003.08082.pdf)
+
+[FedVision: An Online Visual Object Detection Platform Powered by Federated Learning](https://arxiv.org/pdf/2001.06202.pdf)
+
+## Health Care:
+[Multi-Institutional Deep Learning Modeling Without Sharing Patient Data: A Feasibility Study on Brain Tumor Segmentation](https://arxiv.org/pdf/1810.04304.pdf)
+
+[Federated Learning in Distributed Medical Databases: Meta-Analysis of Large-Scale Subcortical Brain Data](https://arxiv.org/pdf/1810.08553.pdf)
+
+[Privacy-Preserving Technology to Help Millions of People: Federated Prediction Model for Stroke Prevention](https://arxiv.org/pdf/2006.10517.pdf)
+
+[A Federated Learning Framework for Healthcare IoT devices](https://arxiv.org/pdf/2005.05083.pdf)
+Keywords: Split Learning + Sparsification
+
+[Federated Transfer Learning for EEG Signal Classification](https://arxiv.org/pdf/2004.12321.pdf)
+
+[The Future of Digital Health with Federated Learning](https://arxiv.org/pdf/2003.08119.pdf)
+
+[Anonymizing Data for Privacy-Preserving Federated Learning. ECAI 2020.](https://arxiv.org/pdf/2002.09096.pdf)
+
+[Federated machine learning with Anonymous Random Hybridization (FeARH) on medical records](https://arxiv.org/pdf/2001.09751.pdf)
+
+[Stratified cross-validation for unbiased and privacy-preserving federated learning](https://arxiv.org/pdf/2001.08090.pdf)
+
+[Multi-site fMRI Analysis Using Privacy-preserving Federated Learning and Domain Adaptation: ABIDE Results](https://arxiv.org/pdf/2001.05647.pdf)
+
+[Learn Electronic Health Records by Fully Decentralized Federated Learning](https://arxiv.org/pdf/1912.01792.pdf)
+
+[Preserving Patient Privacy while Training a Predictive Model of In-hospital Mortality](https://arxiv.org/pdf/1912.00354.pdf)
+
+[Federated Learning for Healthcare Informatics](https://arxiv.org/pdf/1911.06270.pdf)
+
+[Federated and Differentially Private Learning for Electronic Health Records](https://arxiv.org/pdf/1911.05861.pdf)
+
+[A blockchain-orchestrated Federated Learning architecture for healthcare consortia](https://arxiv.org/pdf/1910.12603.pdf)
+
+[Federated Uncertainty-Aware Learning for Distributed Hospital EHR Data](https://arxiv.org/pdf/1910.12191.pdf)
+
+[Stochastic Channel-Based Federated Learning for Medical Data Privacy Preserving](https://arxiv.org/pdf/1910.11160.pdf)
+
+[Differential Privacy-enabled Federated Learning for Sensitive Health Data](https://arxiv.org/pdf/1910.02578.pdf)
+
+[LoAdaBoost: Loss-based AdaBoost federated machine learning with reduced computational complexity on IID and non-IID intensive care data](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0230706)
+
+[Privacy Preserving Stochastic Channel-Based Federated Learning with Neural Network Pruning](https://arxiv.org/pdf/1910.02115.pdf)
+
+[Confederated Machine Learning on Horizontally and Vertically Separated Medical Data for Large-Scale Health System Intelligence](https://arxiv.org/pdf/1910.02109.pdf)
+
+[Privacy-preserving Federated Brain Tumour Segmentation](https://arxiv.org/pdf/1910.00962.pdf)
+
+[HHHFL: Hierarchical Heterogeneous Horizontal Federated Learning for Electroencephalography](https://arxiv.org/pdf/1909.05784.pdf)
+
+[FedHealth: A Federated Transfer Learning Framework for Wearable Healthcare](https://arxiv.org/pdf/1907.09173.pdf)
+
+[Patient Clustering Improves Efficiency of Federated Machine Learning to predict mortality and hospital stay time using distributed Electronic Medical Records](https://arxiv.org/pdf/1903.09296.pdf)
+
+[LoAdaBoost:Loss-Based AdaBoost Federated Machine Learning on medical Data](https://arxiv.org/pdf/1811.12629.pdf)
+
+[FADL:Federated-Autonomous Deep Learning for Distributed Electronic Health Record](https://arxiv.org/pdf/1811.11400.pdf)
+
+
+## Transportation:
+[Federated Learning for Vehicular Networks](https://arxiv.org/pdf/2006.01412.pdf)
+
+[Towards Federated Learning in UAV-Enabled Internet of Vehicles: A Multi-Dimensional Contract-Matching Approach](https://arxiv.org/pdf/2004.03877.pdf)
+
+[Federated Learning Meets Contract Theory: Energy-Efficient Framework for Electric Vehicle Networks](https://arxiv.org/pdf/2004.01828.pdf)
+
+[Beyond privacy regulations: an ethical approach to data usage in transportation. TomTom. 2020-04-01](https://arxiv.org/pdf/2004.00491.pdf)
+
+[Privacy-preserving Traffic Flow Prediction: A Federated Learning Approach](https://arxiv.org/pdf/2003.08725.pdf)
+
+[Communication-Efficient Massive UAV Online Path Control: Federated Learning Meets Mean-Field Game Theory. 2020-03](https://arxiv.org/pdf/2003.04451.pdf)
+
+[FedLoc: Federated Learning Framework for Data-Driven Cooperative Localization and Location Data Processing. 2020-03](https://arxiv.org/pdf/2003.03697.pdf)
+
+[Practical Privacy Preserving POI Recommendation](https://arxiv.org/pdf/2003.02834.pdf)
+
+[Federated Learning for Localization: A Privacy-Preserving Crowdsourcing Method](https://arxiv.org/pdf/2001.01911.pdf)
+
+[Federated Transfer Reinforcement Learning for Autonomous Driving](https://arxiv.org/pdf/1910.06001.pdf)
+
+[Energy Demand Prediction with Federated Learning for Electric Vehicle Networks](https://arxiv.org/pdf/1909.00907.pdf)
+
+[Distributed Federated Learning for Ultra-Reliable Low-Latency Vehicular Communications](https://arxiv.org/pdf/1807.08127.pdf)
+
+[Federated Learning for Ultra-Reliable Low-Latency V2V Communications](https://arxiv.org/pdf/1805.09253.pdf)
+
+[Federated Learning in Vehicular Edge Computing: A Selective Model Aggregation Approach](https://ieeexplore.ieee.org/abstract/document/8964354/)
+
+
+## Recommendation System
+[(*) Federated Multi-view Matrix Factorization for Personalized Recommendations](https://arxiv.org/pdf/2004.04256.pdf)
+
+
+[Robust Federated Recommendation System](https://arxiv.org/pdf/2006.08259.pdf)
+
+[Federated Recommendation System via Differential Privacy](https://arxiv.org/pdf/2005.06670.pdf)
+
+[FedRec: Privacy-Preserving News Recommendation with Federated Learning. MSRA. 2020-03](https://arxiv.org/pdf/2003.09592.pdf)
+
+[Federating Recommendations Using Differentially Private Prototypes](https://arxiv.org/pdf/2003.00602.pdf)
+
+[Meta Matrix Factorization for Federated Rating Predictions](https://arxiv.org/pdf/1910.10086.pdf)
+
+[Federated Hierarchical Hybrid Networks for Clickbait Detection](https://arxiv.org/pdf/1906.00638.pdf)
+
+[Federated Collaborative Filtering for Privacy-Preserving Personalized Recommendation System](https://arxiv.org/pdf/1901.09888.pdf)
+
+## Speech Recognition
+[Training Keyword Spotting Models on Non-IID Data with Federated Learning](https://arxiv.org/pdf/2005.10406.pdf)
+
+## Finance
+[FedCoin: A Peer-to-Peer Payment System for Federated Learning](https://arxiv.org/pdf/2002.11711.pdf)
+
+[Towards Federated Graph Learning for Collaborative Financial Crimes Detection](https://arxiv.org/pdf/1909.12946.pdf)
+
+## Smart City
+[Cloud-based Federated Boosting for Mobile Crowdsensing](https://arxiv.org/pdf/2005.05304.pdf)
+
+[Exploiting Unlabeled Data in Smart Cities using Federated Learning](https://arxiv.org/pdf/2001.04030.pdf)
+
+## Robotics
+[Federated Imitation Learning: A Privacy Considered Imitation Learning Framework for Cloud Robotic Systems with Heterogeneous Sensor Data](https://arxiv.org/pdf/1909.00895.pdf)
+
+[Lifelong Federated Reinforcement Learning: A Learning Architecture for Navigation in Cloud Robotic Systems](https://arxiv.org/pdf/1901.06455.pdf)
+
+## Networking
+[A Federated Learning Approach for Mobile Packet Classification](https://arxiv.org/pdf/1907.13113.pdf)
+
+## Blockchain
+[Blockchained On-Device Federated Learning](https://arxiv.org/pdf/1808.03949.pdf)
+
+[Record and reward federated learning contributions with blockchain](https://mblocklab.com/RecordandReward.pdf)
+
+## Other
+[Boosting Privately: Privacy-Preserving Federated Extreme Boosting for Mobile Crowdsensing](https://arxiv.org/pdf/1907.10218.pdf)
+
+[Self-supervised audio representation learning for mobile devices](https://arxiv.org/pdf/1905.11796.pdf)
+
+[Combining Federated and Active Learning for Communication-efficient Distributed Failure Prediction in Aeronautics](https://arxiv.org/pdf/2001.07504.pdf)
+
+[PMF: A Privacy-preserving Human Mobility Prediction Framework via Federated Learning](https://vonfeng.github.io/files/UbiComp2020_PMF_Final.pdf)
+
+[Federated Multi-task Hierarchical Attention Model for Sensor Analytics](https://arxiv.org/pdf/1905.05142.pdf)
+
+[DÏoT: A Federated Self-learning Anomaly Detection System for IoT](https://arxiv.org/pdf/1804.07474.pdf)
+
+# Benchmark, Dataset and Survey
+
+## Benchmark and Dataset
+
+[The OARF Benchmark Suite: Characterization and Implications for Federated Learning Systems](https://arxiv.org/pdf/2006.07856.pdf)
+
+[Evaluation Framework For Large-scale Federated Learning](https://arxiv.org/pdf/2003.01575.pdf)
+
+[(*) PrivacyFL: A simulator for privacy-preserving and secure federated learning. MIT CSAIL.](https://arxiv.org/pdf/2002.08423.pdf)
+
+[Revocable Federated Learning: A Benchmark of Federated Forest](https://arxiv.org/pdf/1911.03242.pdf)
+
+[Real-World Image Datasets for Federated Learning](https://arxiv.org/pdf/1910.11089.pdf)
+
+[LEAF: A Benchmark for Federated Settings](https://arxiv.org/pdf/1812.01097.pdf)
+
+[Functional Federated Learning in Erlang (ffl-erl)](https://arxiv.org/pdf/1808.08143.pdf)
+
+## Survey
+
+[A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection](https://arxiv.org/pdf/1907.09693.pdf)
+
+Researcher: Bingsheng He, NUS [Qinbin Li, PhD, NUS, HKUST](https://qinbinli.com/files/CV_QB.pdf)
+
+[SECure: A Social and Environmental Certificate for AI Systems](https://arxiv.org/pdf/2006.06217.pdf)
+
+[From Federated Learning to Fog Learning: Towards Large-Scale Distributed Machine Learning in Heterogeneous Wireless Networks](https://arxiv.org/pdf/2006.03594.pdf)
+
+[Federated Learning for 6G Communications: Challenges, Methods, and Future Directions](https://arxiv.org/pdf/2006.02931.pdf)
+
+[A Review of Privacy Preserving Federated Learning for Private IoT Analytics](https://arxiv.org/pdf/2004.11794.pdf)
+
+[Survey of Personalization Techniques for Federated Learning. 2020-03-19](https://arxiv.org/pdf/2003.08673.pdf)
+
+[Threats to Federated Learning: A Survey](https://arxiv.org/pdf/2003.02133.pdf)
+
+[Towards Utilizing Unlabeled Data in Federated Learning: A Survey and Prospective](https://arxiv.org/pdf/2002.11545.pdf)
+
+[Federated Learning for Resource-Constrained IoT Devices: Panoramas and State-of-the-art](https://arxiv.org/pdf/2002.10610.pdf)
+
+[Advances and Open Problems in Federated Learning](https://arxiv.org/pdf/1912.04977.pdf)
+
+[Privacy-Preserving Blockchain Based Federated Learning with Differential Data Sharing](https://arxiv.org/pdf/1912.04859.pdf)
+
+[An Introduction to Communication Efficient Edge Machine Learning](https://arxiv.org/pdf/1912.01554.pdf)
+
+[Federated Learning for Healthcare Informatics](https://arxiv.org/pdf/1911.06270.pdf)
+
+[Federated Learning for Coalition Operations](https://arxiv.org/pdf/1910.06799.pdf)
+
+[Federated Learning in Mobile Edge Networks: A Comprehensive Survey](https://arxiv.org/pdf/1909.11875.pdf)
+
+[Federated Learning: Challenges, Methods, and Future Directions](https://arxiv.org/pdf/1908.07873.pdf)
+
+[A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection](https://arxiv.org/pdf/1907.09693.pdf)
+
+[Federated Machine Learning: Concept and Applications](https://arxiv.org/pdf/1902.04885.pdf)
+
+[No Peek: A Survey of private distributed deep learning](https://arxiv.org/pdf/1812.03288.pdf)
+
+[Communication-Efficient Edge AI: Algorithms and Systems](http://arxiv.org/pdf/2002.09668.pdf)
