Refactor of Curriculum and parameter sampling

Project/Assets/ML-Agents/Examples/WallJump/Scripts/WallJumpAgent.cs

@@ -351,9 +351,7 @@ void ConfigureAgent(int config)
         }
         else
         {
-            var min = m_ResetParams.GetWithDefault("big_wall_min_height", 8);
-            var max = m_ResetParams.GetWithDefault("big_wall_max_height", 8);
-            var height = min + Random.value * (max - min);
+            var height = m_ResetParams.GetWithDefault("big_wall_height", 8);
             localScale = new Vector3(
                 localScale.x,
                 height,

com.unity.ml-agents/CHANGELOG.md

@@ -13,6 +13,10 @@ and this project adheres to
 #### ml-agents / ml-agents-envs / gym-unity (Python)
 - The Parameter Randomization feature has been refactored to enable sampling of new parameters per episode to improve robustness. The
   `resampling-interval` parameter has been removed and the config structure updated. More information [here](https://github.com/Unity-Technologies/ml-agents/blob/master/docs/Training-ML-Agents.md). (#4065)
+- The Parameter Randomization feature has been merged with the Curriculum feature. It is now possible to specify a sampler
+in the lesson of a Curriculum. Curriculum has been refactored and is now specified at the level of the parameter, not the
+behavior. More information
+[here](https://github.com/Unity-Technologies/ml-agents/blob/master/docs/Training-ML-Agents.md).(#4160)
 
 ### Minor Changes
 #### com.unity.ml-agents (C#)

config/ppo/3DBall_randomize.yaml

@@ -24,8 +24,7 @@ behaviors:
     time_horizon: 1000
     summary_freq: 12000
     threaded: true
-
-parameter_randomization:
+environment_parameters:
   mass:
     sampler_type: uniform
     sampler_parameters:

config/ppo/WallJump_curriculum.yaml

@@ -49,20 +49,72 @@ behaviors:
     time_horizon: 128
     summary_freq: 20000
     threaded: true
-
-curriculum:
-  BigWallJump:
-    measure: progress
-    thresholds: [0.1, 0.3, 0.5]
-    min_lesson_length: 100
-    signal_smoothing: true
-    parameters:
-      big_wall_min_height: [0.0, 4.0, 6.0, 8.0]
-      big_wall_max_height: [4.0, 7.0, 8.0, 8.0]
-  SmallWallJump:
-    measure: progress
-    thresholds: [0.1, 0.3, 0.5]
-    min_lesson_length: 100
-    signal_smoothing: true
-    parameters:
-      small_wall_height: [1.5, 2.0, 2.5, 4.0]
+environment_parameters:
+  big_wall_height:
+    curriculum:
+      - name: Lesson0 # The '-' is important as this is a list
+        completion_criteria:
+          measure: progress
+          behavior: BigWallJump
+          signal_smoothing: true
+          min_lesson_length: 100
+          threshold: 0.1
+        value:
+          sampler_type: uniform
+          sampler_parameters:
+            min_value: 0.0
+            max_value: 4.0
+      - name: Lesson1 # This is the start of the second lesson
+        completion_criteria:
+          measure: progress
+          behavior: BigWallJump
+          signal_smoothing: true
+          min_lesson_length: 100
+          threshold: 0.3
+        value:
+          sampler_type: uniform
+          sampler_parameters:
+            min_value: 4.0
+            max_value: 7.0
+      - name: Lesson2
+        completion_criteria:
+          measure: progress
+          behavior: BigWallJump
+          signal_smoothing: true
+          min_lesson_length: 100
+          threshold: 0.5
+        value:
+          sampler_type: uniform
+          sampler_parameters:
+            min_value: 6.0
+            max_value: 8.0
+      - name: Lesson3
+        value: 8.0
+  small_wall_height:
+    curriculum:
+      - name: Lesson0
+        completion_criteria:
+          measure: progress
+          behavior: SmallWallJump
+          signal_smoothing: true
+          min_lesson_length: 100
+          threshold: 0.1
+        value: 1.5
+      - name: Lesson1
+        completion_criteria:
+          measure: progress
+          behavior: SmallWallJump
+          signal_smoothing: true
+          min_lesson_length: 100
+          threshold: 0.3
+        value: 2.0
+      - name: Lesson2
+        completion_criteria:
+          measure: progress
+          behavior: SmallWallJump
+          signal_smoothing: true
+          min_lesson_length: 100
+          threshold: 0.5
+        value: 2.5
+      - name: Lesson3
+        value: 4.0

docs/Migrating.md

@@ -14,7 +14,21 @@ double-check that the versions are in the same. The versions can be found in
 
 # Migrating
 
-## Migrating from Release 1 to latest
+## Migrating from Release 3 to latest
+
+### Important changes
+- The Parameter Randomization feature has been merged with the Curriculum feature. It is now possible to specify a sampler
+in the lesson of a Curriculum. Curriculum has been refactored and is now specified at the level of the parameter, not the
+behavior. More information
+[here](https://github.com/Unity-Technologies/ml-agents/blob/master/docs/Training-ML-Agents.md).(#4160)
+
+### Steps to Migrate
+- The configuration format for curriculum and parameter randomization has changed. To upgrade your configuration files,
+an upgrade script has been provided. Run `python -m mlagents.trainers.upgrade_config -h` to see the script usage. Note that you will have had to upgrade to/install the current version of ML-Agents before running the script. To update manually:
+  - If your config file used a `parameter_randomization` section, rename that section to `environment_parameters`
+  - If your config file used a `curriculum` section, you will need to rewrite your curriculum with this [format](Training-ML-Agents.md#curriculum).
+
+## Migrating from Release 1 to Release 3
 
 ### Important changes
 - Training artifacts (trained models, summaries) are now found under `results/`