Got the walker to walk different based on diversity setting

config/ppo/Pyramids.yaml

       hidden_units: 512
       num_layers: 2
       vis_encode_type: simple
+      goal_conditioning_type: none
-      curiosity:
+      # curiosity:
+      #   gamma: 0.99
+      #   strength: 0.02
+      #   network_settings:
+      #     hidden_units: 256
+      #   learning_rate: 0.0003
+      diverse:
-        strength: 0.02
-        network_settings:
-          hidden_units: 256
+        strength: 0.1
         learning_rate: 0.0003
     keep_checkpoints: 5
     max_steps: 10000000

config/ppo/Walker.yaml

       hidden_units: 512
       num_layers: 3
       vis_encode_type: simple
+      goal_conditioning_type: none
+      diverse:
+        gamma: 0.99
+        strength: 0.1
+        network_settings:
+          normalize: false
+          hidden_units: 512
+          num_layers: 3
+          vis_encode_type: simple
+          goal_conditioning_type: none
     keep_checkpoints: 5
     max_steps: 30000000
     time_horizon: 1000

config/ppo/GridWorld.yaml

       normalize: false
       hidden_units: 128
       num_layers: 1
-      vis_encode_type: simple
+      vis_encode_type: fully_connected
     reward_signals:
       extrinsic:
         gamma: 0.9

config/sac/Pyramids.yaml

       hidden_units: 512
       num_layers: 3
       vis_encode_type: simple
+      goal_conditioning_type: none
-      gail:
-        gamma: 0.99
-        strength: 0.01
-        learning_rate: 0.0003
-        use_actions: true
-        use_vail: false
-        demo_path: Project/Assets/ML-Agents/Examples/Pyramids/Demos/ExpertPyramid.demo
+      # gail:
+      #   gamma: 0.99
+      #   strength: 0.01
+      #   learning_rate: 0.0003
+      #   use_actions: true
+      #   use_vail: false
+      #   demo_path: Project/Assets/ML-Agents/Examples/Pyramids/Demos/ExpertPyramid.demo
-    max_steps: 3000000
+    max_steps: 30000000
     time_horizon: 128
     summary_freq: 30000

config/sac/Walker.yaml

       hidden_units: 256
       num_layers: 3
       vis_encode_type: simple
+      goal_conditioning_type: none
-        strength: 1.0
+        strength: 0.1
-    max_steps: 15000000
+    max_steps: 150000000
     time_horizon: 1000
     summary_freq: 30000

com.unity.ml-agents/CHANGELOG.md

 sizes and will need to be retrained. (#5181)
 - The `AbstractBoard` class for integration with Match-3 games was changed to make it easier to support boards with
 different sizes using the same model. For a summary of the interface changes, please see the Migration Guide. (##5189)
-- Updated the Barracuda package to version `1.3.3-preview`(#5236)
+- Updated the Barracuda package to version `1.4.0-preview`(#5236)
 - `GridSensor` has been refactored and moved to main package, with changes to both sensor interfaces and behaviors.
 Exsisting GridSensor created by extension package will not work in newer version. Previously trained models will
 need to be retrained. Please see the Migration Guide for more details. (#5256)

com.unity.ml-agents/package.json

   "unity": "2019.4",
   "description": "Use state-of-the-art machine learning to create intelligent character behaviors in any Unity environment (games, robotics, film, etc.).",
   "dependencies": {
-    "com.unity.barracuda": "1.3.3-preview",
+    "com.unity.barracuda": "1.4.0-preview",
     "com.unity.modules.imageconversion": "1.0.0",
     "com.unity.modules.jsonserialize": "1.0.0"
   }

Project/Assets/ML-Agents/Examples/PushBlock/Prefabs/PushBlockArea.prefab

   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   - component: {fileID: 114451319691753174}
   - component: {fileID: 8964598783836598940}
   - component: {fileID: 4081319787948195948}
+  - component: {fileID: 3604832710400578108}
   m_Layer: 0
   m_Name: Agent
   m_TagString: agent
   m_Name: 
   m_EditorClassIdentifier: 
   debugCommandLineOverride: 
+--- !u!114 &3604832710400578108
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
+  m_GameObject: {fileID: 1489716781518988}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 38b7cc1f5819445aa85e9a9b054552dc, type: 3}
+  m_Name: 
+  m_EditorClassIdentifier: 
+  m_SensorName: VectorSensor
+  m_ObservationSize: 8
+  m_ObservationType: 1
 --- !u!1 &1500989011945850
 GameObject:
   m_ObjectHideFlags: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_ClearFlags: 2
   m_BackGroundColor: {r: 0.46666667, g: 0.5647059, b: 0.60784316, a: 1}
   m_projectionMatrixMode: 1
+  m_GateFitMode: 2
+  m_FOVAxisMode: 0
-  m_GateFitMode: 2
   m_FocalLength: 50
   m_NormalizedViewPortRect:
     serializedVersion: 2
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0

Project/Assets/ML-Agents/Examples/PushBlock/Scripts/PushAgentBasic.cs

 using UnityEngine;
 using Unity.MLAgents;
 using Unity.MLAgents.Actuators;
+using Unity.MLAgents.Sensors;
 
 public class PushAgentBasic : Agent
 {
         m_ResetParams = Academy.Instance.EnvironmentParameters;
 
         SetResetParameters();
+        GetComponent<VectorSensorComponent>().CreateSensors();
+        m_DiversitySettingSensor = GetComponent<VectorSensorComponent>();
+    }
+
+    VectorSensorComponent m_DiversitySettingSensor;
+    public int m_DiversitySetting = 0;
+
+
+    /// <summary>
+    /// Loop over body parts to add them to observation.
+    /// </summary>
+    public override void CollectObservations(VectorSensor sensor)
+    {
+        m_DiversitySettingSensor.GetSensor().Reset();
+        m_DiversitySettingSensor.GetSensor().AddOneHotObservation(m_DiversitySetting, 8);
     }
 
     /// <summary>
         m_AgentRb.angularVelocity = Vector3.zero;
 
         SetResetParameters();
+        m_DiversitySetting = Random.Range(0, 8);
     }
 
     public void SetGroundMaterialFriction()

Project/Assets/ML-Agents/Examples/Pyramids/Prefabs/AreaPB.prefab

   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   - component: {fileID: 5767481171805996936}
   - component: {fileID: 4725417187860315718}
   - component: {fileID: 6474351450651730614}
+  - component: {fileID: 5328107234309071792}
   m_Layer: 0
   m_Name: Agent
   m_TagString: agent
   m_Name: 
   m_EditorClassIdentifier: 
   debugCommandLineOverride: 
+--- !u!114 &5328107234309071792
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
+  m_GameObject: {fileID: 1131043459059966}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 38b7cc1f5819445aa85e9a9b054552dc, type: 3}
+  m_Name: 
+  m_EditorClassIdentifier: 
+  m_SensorName: VectorSensor
+  m_ObservationSize: 8
+  m_ObservationType: 1
 --- !u!1 &1148882946833254
 GameObject:
   m_ObjectHideFlags: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_ClearFlags: 2
   m_BackGroundColor: {r: 0.46666667, g: 0.5647059, b: 0.60784316, a: 1}
   m_projectionMatrixMode: 1
+  m_GateFitMode: 2
+  m_FOVAxisMode: 0
-  m_GateFitMode: 2
   m_FocalLength: 50
   m_NormalizedViewPortRect:
     serializedVersion: 2
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0

Project/Assets/ML-Agents/Examples/Pyramids/Scripts/PyramidAgent.cs

     public GameObject areaSwitch;
     public bool useVectorObs;
 
+
+    VectorSensorComponent m_DiversitySettingSensor;
+    public int m_DiversitySetting = 0;
+
+                GetComponent<VectorSensorComponent>().CreateSensors();
+        m_DiversitySettingSensor = GetComponent<VectorSensorComponent>();
     }
 
     public override void CollectObservations(VectorSensor sensor)
             sensor.AddObservation(m_SwitchLogic.GetState());
             sensor.AddObservation(transform.InverseTransformDirection(m_AgentRb.velocity));
         }
+
+        m_DiversitySettingSensor.GetSensor().Reset();
+        m_DiversitySettingSensor.GetSensor().AddOneHotObservation(m_DiversitySetting, 8);
     }
 
     public void MoveAgent(ActionSegment<int> act)
         m_MyArea.CreateStonePyramid(1, items[6]);
         m_MyArea.CreateStonePyramid(1, items[7]);
         m_MyArea.CreateStonePyramid(1, items[8]);
+
+        m_DiversitySetting = Random.Range(0, 8);
     }
 
     void OnCollisionEnter(Collision collision)

Project/Assets/ML-Agents/Examples/Walker/Scripts/WalkerAgent.cs

         m_ResetParams = Academy.Instance.EnvironmentParameters;
 
         SetResetParameters();
+        GetComponent<VectorSensorComponent>().CreateSensors();
+        m_DiversitySettingSensor = GetComponent<VectorSensorComponent>();
+
+    VectorSensorComponent m_DiversitySettingSensor;
+    public int m_DiversitySetting = 0;
 
     /// <summary>
     /// Loop over body parts and reset them to initial conditions.
             randomizeWalkSpeedEachEpisode ? Random.Range(0.1f, m_maxWalkingSpeed) : MTargetWalkingSpeed;
 
         SetResetParameters();
+
+        m_DiversitySetting = Random.Range(0, 8);
     }
 
     /// <summary>
     /// </summary>
     public override void CollectObservations(VectorSensor sensor)
     {
+        m_DiversitySettingSensor.GetSensor().Reset();
+        m_DiversitySettingSensor.GetSensor().AddOneHotObservation(m_DiversitySetting, 8);
+
         var cubeForward = m_OrientationCube.transform.forward;
 
         //velocity we want to match
     public void TouchedTarget()
     {
         AddReward(1f);
+        //Set our goal walking speed
+        MTargetWalkingSpeed =
+            randomizeWalkSpeedEachEpisode ? Random.Range(0.1f, m_maxWalkingSpeed) : MTargetWalkingSpeed;
+
+        SetResetParameters();
+
+        m_DiversitySetting = Random.Range(0, 8);
     }
 
     public void SetTorsoMass()

Project/Assets/ML-Agents/Examples/Walker/Prefabs/Platforms/Platform.prefab

     m_Modifications:
     - target: {fileID: 6902107422946006027, guid: f0d7741d9e06247f6843b921a206b978,
         type: 3}
+      propertyPath: m_RootOrder
+      value: 0
+      objectReference: {fileID: 0}
+    - target: {fileID: 6902107422946006027, guid: f0d7741d9e06247f6843b921a206b978,
+        type: 3}
       propertyPath: m_LocalPosition.x
       value: 0
       objectReference: {fileID: 0}
       objectReference: {fileID: 0}
     - target: {fileID: 6902107422946006027, guid: f0d7741d9e06247f6843b921a206b978,
         type: 3}
+      propertyPath: m_LocalRotation.w
+      value: 1
+      objectReference: {fileID: 0}
+    - target: {fileID: 6902107422946006027, guid: f0d7741d9e06247f6843b921a206b978,
+        type: 3}
       propertyPath: m_LocalRotation.x
       value: -0
       objectReference: {fileID: 0}
       objectReference: {fileID: 0}
     - target: {fileID: 6902107422946006027, guid: f0d7741d9e06247f6843b921a206b978,
         type: 3}
-      propertyPath: m_LocalRotation.w
-      value: 1
-      objectReference: {fileID: 0}
-    - target: {fileID: 6902107422946006027, guid: f0d7741d9e06247f6843b921a206b978,
-        type: 3}
-      propertyPath: m_RootOrder
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 6902107422946006027, guid: f0d7741d9e06247f6843b921a206b978,
-        type: 3}
       propertyPath: m_LocalEulerAnglesHint.x
       value: 0
       objectReference: {fileID: 0}
     m_Modifications:
     - target: {fileID: 3839136118347789758, guid: 46734abd0de454192b407379c6a4ab8d,
         type: 3}
+      propertyPath: m_RootOrder
+      value: 2
+      objectReference: {fileID: 0}
+    - target: {fileID: 3839136118347789758, guid: 46734abd0de454192b407379c6a4ab8d,
+        type: 3}
       propertyPath: m_LocalPosition.x
       value: 0
       objectReference: {fileID: 0}
       objectReference: {fileID: 0}
     - target: {fileID: 3839136118347789758, guid: 46734abd0de454192b407379c6a4ab8d,
         type: 3}
+      propertyPath: m_LocalRotation.w
+      value: 1
+      objectReference: {fileID: 0}
+    - target: {fileID: 3839136118347789758, guid: 46734abd0de454192b407379c6a4ab8d,
+        type: 3}
       propertyPath: m_LocalRotation.x
       value: 0
       objectReference: {fileID: 0}
       objectReference: {fileID: 0}
     - target: {fileID: 3839136118347789758, guid: 46734abd0de454192b407379c6a4ab8d,
         type: 3}
-      propertyPath: m_LocalRotation.w
-      value: 1
-      objectReference: {fileID: 0}
-    - target: {fileID: 3839136118347789758, guid: 46734abd0de454192b407379c6a4ab8d,
-        type: 3}
-      propertyPath: m_RootOrder
-      value: 2
-      objectReference: {fileID: 0}
-    - target: {fileID: 3839136118347789758, guid: 46734abd0de454192b407379c6a4ab8d,
-        type: 3}
       propertyPath: m_LocalEulerAnglesHint.x
       value: 0
       objectReference: {fileID: 0}
       objectReference: {fileID: 0}
     - target: {fileID: 895268871377934297, guid: 765582efd9dda46ed98564603316353f,
         type: 3}
+      propertyPath: m_Model
+      value: 
+      objectReference: {fileID: 5022602860645237092, guid: a043eafd232b84def8999a7288cc791b,
+        type: 3}
+    - target: {fileID: 895268871377934297, guid: 765582efd9dda46ed98564603316353f,
+        type: 3}
-    - target: {fileID: 895268871377934297, guid: 765582efd9dda46ed98564603316353f,
+    - target: {fileID: 895268871377934298, guid: 765582efd9dda46ed98564603316353f,
-      propertyPath: m_Model
-      value: 
-      objectReference: {fileID: 11400000, guid: 205590a7f0a844b24b82b7f8355a1529,
-        type: 3}
+      propertyPath: m_RootOrder
+      value: 3
+      objectReference: {fileID: 0}
     - target: {fileID: 895268871377934298, guid: 765582efd9dda46ed98564603316353f,
         type: 3}
       propertyPath: m_LocalPosition.x
       objectReference: {fileID: 0}
     - target: {fileID: 895268871377934298, guid: 765582efd9dda46ed98564603316353f,
         type: 3}
+      propertyPath: m_LocalRotation.w
+      value: 1
+      objectReference: {fileID: 0}
+    - target: {fileID: 895268871377934298, guid: 765582efd9dda46ed98564603316353f,
+        type: 3}
       propertyPath: m_LocalRotation.x
       value: 0
       objectReference: {fileID: 0}
         type: 3}
       propertyPath: m_LocalRotation.z
       value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 895268871377934298, guid: 765582efd9dda46ed98564603316353f,
-        type: 3}
-      propertyPath: m_LocalRotation.w
-      value: 1
-      objectReference: {fileID: 0}
-    - target: {fileID: 895268871377934298, guid: 765582efd9dda46ed98564603316353f,
-        type: 3}
-      propertyPath: m_RootOrder
-      value: 3
       objectReference: {fileID: 0}
     - target: {fileID: 895268871377934298, guid: 765582efd9dda46ed98564603316353f,
         type: 3}

Project/Assets/ML-Agents/Examples/Walker/Prefabs/Ragdoll/WalkerRagdoll.prefab

   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_EditorClassIdentifier: 
   agent: {fileID: 0}
   agentDoneOnGroundContact: 1
-  penalizeGroundContact: 1
+  penalizeGroundContact: 0
   groundContactPenalty: -1
   touchingGround: 0
 --- !u!1 &895268871289741235
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   - component: {fileID: 895268871377934303}
   - component: {fileID: 895268871377934302}
   - component: {fileID: 895268871377934301}
+  - component: {fileID: 5678764055635236588}
   m_Layer: 0
   m_Name: WalkerRagdoll
   m_TagString: Untagged
     VectorActionDescriptions: []
     VectorActionSpaceType: 1
     hasUpgradedBrainParametersWithActionSpec: 1
-  m_Model: {fileID: 11400000, guid: f598eaeeef9f94691989a2cfaaafb565, type: 3}
+  m_Model: {fileID: 5022602860645237092, guid: a043eafd232b84def8999a7288cc791b, type: 3}
   m_InferenceDevice: 2
   m_BehaviorType: 0
   m_BehaviorName: WalkerDynamic
   armR: {fileID: 7933235355057813930}
   forearmR: {fileID: 7933235353195701980}
   handR: {fileID: 7933235354616748502}
+  m_DiversitySetting: 0
 --- !u!114 &895268871377934303
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_Name: 
   m_EditorClassIdentifier: 
   debugCommandLineOverride: 
+--- !u!114 &5678764055635236588
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
+  m_GameObject: {fileID: 895268871377934275}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 38b7cc1f5819445aa85e9a9b054552dc, type: 3}
+  m_Name: 
+  m_EditorClassIdentifier: 
+  m_SensorName: DiversitySetting
+  m_ObservationSize: 8
+  m_ObservationType: 1
 --- !u!1 &895268871382313704
 GameObject:
   m_ObjectHideFlags: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 1
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RayTracingMode: 2
   m_RenderingLayerMask: 1
   m_RendererPriority: 0
   m_Materials:
   m_ProbeAnchor: {fileID: 0}
   m_LightProbeVolumeOverride: {fileID: 0}
   m_ScaleInLightmap: 1
+  m_ReceiveGI: 1
   m_PreserveUVs: 0
   m_IgnoreNormalsForChartDetection: 0
   m_ImportantGI: 0
   m_EditorClassIdentifier: 
   agent: {fileID: 0}
   agentDoneOnGroundContact: 1
-  penalizeGroundContact: 1
+  penalizeGroundContact: 0
   groundContactPenalty: -1
   touchingGround: 0
 --- !u!136 &7933235353030744117
   m_EditorClassIdentifier: 
   agent: {fileID: 0}
   agentDoneOnGroundContact: 1
-  penalizeGroundContact: 1
+  penalizeGroundContact: 0
   groundContactPenalty: -1
   touchingGround: 0
 --- !u!135 &7933235353041637845
   m_EditorClassIdentifier: 
   agent: {fileID: 0}
   agentDoneOnGroundContact: 1
-  penalizeGroundContact: 1
+  penalizeGroundContact: 0
   groundContactPenalty: -1
   touchingGround: 0
 --- !u!136 &7933235353195701957
   m_EditorClassIdentifier: 
   agent: {fileID: 0}
   agentDoneOnGroundContact: 1
-  penalizeGroundContact: 1
+  penalizeGroundContact: 0
   groundContactPenalty: -1
   touchingGround: 0
 --- !u!136 &7933235353228551178
   m_EditorClassIdentifier: 
   agent: {fileID: 0}
   agentDoneOnGroundContact: 1
-  penalizeGroundContact: 1
+  penalizeGroundContact: 0
   groundContactPenalty: -1
   touchingGround: 0
 --- !u!136 &7933235353240438144
   m_EditorClassIdentifier: 
   agent: {fileID: 0}
   agentDoneOnGroundContact: 1
-  penalizeGroundContact: 1
+  penalizeGroundContact: 0
   groundContactPenalty: -1
   touchingGround: 0
 --- !u!136 &7933235353713167634
   m_EditorClassIdentifier: 
   agent: {fileID: 0}
   agentDoneOnGroundContact: 1
-  penalizeGroundContact: 1
+  penalizeGroundContact: 0
   groundContactPenalty: -1
   touchingGround: 0
 --- !u!135 &7933235354074184676
   m_EditorClassIdentifier: 
   agent: {fileID: 0}
   agentDoneOnGroundContact: 1
-  penalizeGroundContact: 1
+  penalizeGroundContact: 0
   groundContactPenalty: -1
   touchingGround: 0
 --- !u!135 &7933235354616748520
   m_EditorClassIdentifier: 
   agent: {fileID: 0}
   agentDoneOnGroundContact: 1
-  penalizeGroundContact: 1
+  penalizeGroundContact: 0
   groundContactPenalty: -1
   touchingGround: 0
 --- !u!136 &7933235354652902042
   m_EditorClassIdentifier: 
   agent: {fileID: 0}
   agentDoneOnGroundContact: 1
-  penalizeGroundContact: 1
+  penalizeGroundContact: 0
   groundContactPenalty: -1
   touchingGround: 0
 --- !u!136 &7933235354845945040
   m_EditorClassIdentifier: 
   agent: {fileID: 0}
   agentDoneOnGroundContact: 1
-  penalizeGroundContact: 1
+  penalizeGroundContact: 0
   groundContactPenalty: -1
   touchingGround: 0
 --- !u!136 &7933235355057813907
       objectReference: {fileID: 0}
     - target: {fileID: 2591864627249999504, guid: 72f745913c5a34df5aaadd5c1f0024cb,
         type: 3}
+      propertyPath: m_RootOrder
+      value: 2
+      objectReference: {fileID: 0}
+    - target: {fileID: 2591864627249999504, guid: 72f745913c5a34df5aaadd5c1f0024cb,
+        type: 3}
       propertyPath: m_LocalPosition.x
       value: 0
       objectReference: {fileID: 0}
       objectReference: {fileID: 0}
     - target: {fileID: 2591864627249999504, guid: 72f745913c5a34df5aaadd5c1f0024cb,
         type: 3}
+      propertyPath: m_LocalRotation.w
+      value: 1
+      objectReference: {fileID: 0}
+    - target: {fileID: 2591864627249999504, guid: 72f745913c5a34df5aaadd5c1f0024cb,
+        type: 3}
       propertyPath: m_LocalRotation.x
       value: 0
       objectReference: {fileID: 0}
         type: 3}
       propertyPath: m_LocalRotation.z
       value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 2591864627249999504, guid: 72f745913c5a34df5aaadd5c1f0024cb,
-        type: 3}
-      propertyPath: m_LocalRotation.w
-      value: 1
-      objectReference: {fileID: 0}
-    - target: {fileID: 2591864627249999504, guid: 72f745913c5a34df5aaadd5c1f0024cb,
-        type: 3}
-      propertyPath: m_RootOrder
-      value: 2
       objectReference: {fileID: 0}
     - target: {fileID: 2591864627249999504, guid: 72f745913c5a34df5aaadd5c1f0024cb,
         type: 3}

Project/Assets/ML-Agents/Examples/Walker/Scenes/Walker.unity

   m_ReflectionIntensity: 1
   m_CustomReflection: {fileID: 0}
   m_Sun: {fileID: 0}
-  m_IndirectSpecularColor: {r: 0.4497121, g: 0.49977785, b: 0.57563704, a: 1}
+  m_IndirectSpecularColor: {r: 0.44971168, g: 0.4997775, b: 0.57563686, a: 1}
   m_UseRadianceAmbientProbe: 0
 --- !u!157 &3
 LightmapSettings:
   m_Modification:
     m_TransformParent: {fileID: 0}
     m_Modifications:
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_TargetWalkingSpeed
+      value: 5
+      objectReference: {fileID: 0}
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: randomizeWalkSpeedEachEpisode
+      value: 0
+      objectReference: {fileID: 0}
     - target: {fileID: 1076680649171575083, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_ConnectedAnchor.x
       propertyPath: m_Name
       value: PlatformDynamicTarget (4)
       objectReference: {fileID: 0}
+    - target: {fileID: 6713178126238440196, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_IsActive
+      value: 1
+      objectReference: {fileID: 0}
     - target: {fileID: 6718791046026642300, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_RootOrder
       propertyPath: m_LocalEulerAnglesHint.z
       value: 0
       objectReference: {fileID: 0}
+    - target: {fileID: 7819659174506835736, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_Model
+      value: 
+      objectReference: {fileID: 5022602860645237092, guid: 371ea8d3fe4bf4cde9da72b71d914c52,
+        type: 3}
     m_RemovedComponents: []
   m_SourcePrefab: {fileID: 100100000, guid: 84359146bf7af47e58c229d877e801d7, type: 3}
 --- !u!1001 &193531851
   m_Modification:
     m_TransformParent: {fileID: 0}
     m_Modifications:
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_TargetWalkingSpeed
+      value: 5
+      objectReference: {fileID: 0}
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: randomizeWalkSpeedEachEpisode
+      value: 0
+      objectReference: {fileID: 0}
     - target: {fileID: 1076680649171575083, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_ConnectedAnchor.x
       propertyPath: m_Name
       value: PlatformDynamicTarget (2)
       objectReference: {fileID: 0}
+    - target: {fileID: 6713178126238440196, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_IsActive
+      value: 1
+      objectReference: {fileID: 0}
     - target: {fileID: 6718791046026642300, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_RootOrder
       propertyPath: m_LocalEulerAnglesHint.z
       value: 0
       objectReference: {fileID: 0}
+    - target: {fileID: 7819659174506835736, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_Model
+      value: 
+      objectReference: {fileID: 5022602860645237092, guid: 371ea8d3fe4bf4cde9da72b71d914c52,
+        type: 3}
     m_RemovedComponents: []
   m_SourcePrefab: {fileID: 100100000, guid: 84359146bf7af47e58c229d877e801d7, type: 3}
 --- !u!1001 &476292838
   m_Modification:
     m_TransformParent: {fileID: 0}
     m_Modifications:
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_TargetWalkingSpeed
+      value: 5
+      objectReference: {fileID: 0}
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: randomizeWalkSpeedEachEpisode
+      value: 0
+      objectReference: {fileID: 0}
     - target: {fileID: 1076680649171575083, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_ConnectedAnchor.x
       propertyPath: m_Name
       value: PlatformDynamicTarget (6)
       objectReference: {fileID: 0}
+    - target: {fileID: 6713178126238440196, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_IsActive
+      value: 1
+      objectReference: {fileID: 0}
     - target: {fileID: 6718791046026642300, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_RootOrder
       propertyPath: m_LocalEulerAnglesHint.z
       value: 0
       objectReference: {fileID: 0}
+    - target: {fileID: 7819659174506835736, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_Model
+      value: 
+      objectReference: {fileID: 5022602860645237092, guid: 371ea8d3fe4bf4cde9da72b71d914c52,
+        type: 3}
     m_RemovedComponents: []
   m_SourcePrefab: {fileID: 100100000, guid: 84359146bf7af47e58c229d877e801d7, type: 3}
 --- !u!1 &781961355
   m_Modification:
     m_TransformParent: {fileID: 0}
     m_Modifications:
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_TargetWalkingSpeed
+      value: 5
+      objectReference: {fileID: 0}
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: randomizeWalkSpeedEachEpisode
+      value: 0
+      objectReference: {fileID: 0}
     - target: {fileID: 1076680649171575083, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_ConnectedAnchor.x
       propertyPath: m_Name
       value: PlatformDynamicTarget (9)
       objectReference: {fileID: 0}
+    - target: {fileID: 6713178126238440196, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_IsActive
+      value: 1
+      objectReference: {fileID: 0}
     - target: {fileID: 6718791046026642300, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_RootOrder
       propertyPath: m_LocalEulerAnglesHint.z
       value: 0
       objectReference: {fileID: 0}
+    - target: {fileID: 7819659174506835736, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_Model
+      value: 
+      objectReference: {fileID: 5022602860645237092, guid: 371ea8d3fe4bf4cde9da72b71d914c52,
+        type: 3}
     m_RemovedComponents: []
   m_SourcePrefab: {fileID: 100100000, guid: 84359146bf7af47e58c229d877e801d7, type: 3}
 --- !u!1001 &1062792380
   m_Modification:
     m_TransformParent: {fileID: 0}
     m_Modifications:
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_TargetWalkingSpeed
+      value: 5
+      objectReference: {fileID: 0}
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: randomizeWalkSpeedEachEpisode
+      value: 0
+      objectReference: {fileID: 0}
     - target: {fileID: 1076680649171575083, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_ConnectedAnchor.x
       propertyPath: m_Name
       value: PlatformDynamicTarget (8)
       objectReference: {fileID: 0}
+    - target: {fileID: 6713178126238440196, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_IsActive
+      value: 1
+      objectReference: {fileID: 0}
     - target: {fileID: 6718791046026642300, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_RootOrder
       propertyPath: m_LocalEulerAnglesHint.z
       value: 0
       objectReference: {fileID: 0}
+    - target: {fileID: 7819659174506835736, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_Model
+      value: 
+      objectReference: {fileID: 5022602860645237092, guid: 371ea8d3fe4bf4cde9da72b71d914c52,
+        type: 3}
     m_RemovedComponents: []
   m_SourcePrefab: {fileID: 100100000, guid: 84359146bf7af47e58c229d877e801d7, type: 3}
 --- !u!1001 &1071024415
   m_Modification:
     m_TransformParent: {fileID: 0}
     m_Modifications:
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_TargetWalkingSpeed
+      value: 5
+      objectReference: {fileID: 0}
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: randomizeWalkSpeedEachEpisode
+      value: 0
+      objectReference: {fileID: 0}
     - target: {fileID: 1076680649171575083, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_ConnectedAnchor.x
       propertyPath: m_Name
       value: PlatformDynamicTarget (3)
       objectReference: {fileID: 0}
+    - target: {fileID: 6713178126238440196, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_IsActive
+      value: 1
+      objectReference: {fileID: 0}
     - target: {fileID: 6718791046026642300, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_RootOrder
       propertyPath: m_LocalEulerAnglesHint.z
       value: 0
       objectReference: {fileID: 0}
+    - target: {fileID: 7819659174506835736, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_Model
+      value: 
+      objectReference: {fileID: 5022602860645237092, guid: 371ea8d3fe4bf4cde9da72b71d914c52,
+        type: 3}
     m_RemovedComponents: []
   m_SourcePrefab: {fileID: 100100000, guid: 84359146bf7af47e58c229d877e801d7, type: 3}
 --- !u!1001 &1237883783
   m_Modification:
     m_TransformParent: {fileID: 0}
     m_Modifications:
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_TargetWalkingSpeed
+      value: 5
+      objectReference: {fileID: 0}
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: randomizeWalkSpeedEachEpisode
+      value: 0
+      objectReference: {fileID: 0}
+      objectReference: {fileID: 0}
+    - target: {fileID: 6713178126238440196, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_IsActive
+      value: 1
       objectReference: {fileID: 0}
     - target: {fileID: 6718791046026642300, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_LocalEulerAnglesHint.z
       value: 0
       objectReference: {fileID: 0}
+    - target: {fileID: 7819659174506835736, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_Model
+      value: 
+      objectReference: {fileID: 5022602860645237092, guid: 371ea8d3fe4bf4cde9da72b71d914c52,
+        type: 3}
     m_RemovedComponents: []
   m_SourcePrefab: {fileID: 100100000, guid: 84359146bf7af47e58c229d877e801d7, type: 3}
 --- !u!1 &1392866527
   m_Modification:
     m_TransformParent: {fileID: 0}
     m_Modifications:
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_TargetWalkingSpeed
+      value: 5
+      objectReference: {fileID: 0}
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: randomizeWalkSpeedEachEpisode
+      value: 0
+      objectReference: {fileID: 0}
     - target: {fileID: 1076680649171575083, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_ConnectedAnchor.x
       propertyPath: m_Name
       value: PlatformDynamicTarget (5)
       objectReference: {fileID: 0}
+    - target: {fileID: 6713178126238440196, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_IsActive
+      value: 1
+      objectReference: {fileID: 0}
     - target: {fileID: 6718791046026642300, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_RootOrder
       propertyPath: m_LocalEulerAnglesHint.z
       value: 0
       objectReference: {fileID: 0}
+    - target: {fileID: 7819659174506835736, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_Model
+      value: 
+      objectReference: {fileID: 5022602860645237092, guid: 371ea8d3fe4bf4cde9da72b71d914c52,
+        type: 3}
     m_RemovedComponents: []
   m_SourcePrefab: {fileID: 100100000, guid: 84359146bf7af47e58c229d877e801d7, type: 3}
 --- !u!1001 &1481808307
   m_Modification:
     m_TransformParent: {fileID: 0}
     m_Modifications:
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_TargetWalkingSpeed
+      value: 5
+      objectReference: {fileID: 0}
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: randomizeWalkSpeedEachEpisode
+      value: 0
+      objectReference: {fileID: 0}
     - target: {fileID: 1076680649171575083, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_ConnectedAnchor.x
       propertyPath: m_Name
       value: PlatformDynamicTarget (7)
       objectReference: {fileID: 0}
+    - target: {fileID: 6713178126238440196, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_IsActive
+      value: 1
+      objectReference: {fileID: 0}
     - target: {fileID: 6718791046026642300, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_RootOrder
       propertyPath: m_LocalEulerAnglesHint.z
       value: 0
       objectReference: {fileID: 0}
+    - target: {fileID: 7819659174506835736, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_Model
+      value: 
+      objectReference: {fileID: 5022602860645237092, guid: 371ea8d3fe4bf4cde9da72b71d914c52,
+        type: 3}
     m_RemovedComponents: []
   m_SourcePrefab: {fileID: 100100000, guid: 84359146bf7af47e58c229d877e801d7, type: 3}
 --- !u!1001 &2709362470382547191
   m_Modification:
     m_TransformParent: {fileID: 0}
     m_Modifications:
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_TargetWalkingSpeed
+      value: 5
+      objectReference: {fileID: 0}
+    - target: {fileID: 443215535079112744, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: randomizeWalkSpeedEachEpisode
+      value: 0
+      objectReference: {fileID: 0}
     - target: {fileID: 6713178126238440196, guid: 84359146bf7af47e58c229d877e801d7,
         type: 3}
       propertyPath: m_Name
       propertyPath: m_LocalEulerAnglesHint.z
       value: 0
       objectReference: {fileID: 0}
+    - target: {fileID: 7819659174506835736, guid: 84359146bf7af47e58c229d877e801d7,
+        type: 3}
+      propertyPath: m_Model
+      value: 
+      objectReference: {fileID: 5022602860645237092, guid: 371ea8d3fe4bf4cde9da72b71d914c52,
+        type: 3}
     m_RemovedComponents: []
   m_SourcePrefab: {fileID: 100100000, guid: 84359146bf7af47e58c229d877e801d7, type: 3}

ml-agents/mlagents/trainers/settings.py

     GAIL: str = "gail"
     CURIOSITY: str = "curiosity"
     RND: str = "rnd"
+    DIVERSE: str = "diverse"
 
     def to_settings(self) -> type:
         _mapping = {
             RewardSignalType.RND: RNDSettings,
+            RewardSignalType.DIVERSE: DiverseSettings,
         }
         return _mapping[self]
 
 
 @attr.s(auto_attribs=True)
 class RNDSettings(RewardSignalSettings):
+    learning_rate: float = 1e-4
+    encoding_size: Optional[int] = None
+
+
+@attr.s(auto_attribs=True)
+class DiverseSettings(RewardSignalSettings):
     learning_rate: float = 1e-4
     encoding_size: Optional[int] = None
 

ml-agents/mlagents/trainers/torch/components/reward_providers/__init__.py

 from mlagents.trainers.torch.components.reward_providers.rnd_reward_provider import (  # noqa F401
     RNDRewardProvider,
 )
+from mlagents.trainers.torch.components.reward_providers.diverse_reward_provider import (  # noqa F401
+    DiverseRewardProvider,
+)
 from mlagents.trainers.torch.components.reward_providers.reward_provider_factory import (  # noqa F401
     create_reward_provider,
 )

ml-agents/mlagents/trainers/torch/components/reward_providers/reward_provider_factory.py

 from mlagents.trainers.torch.components.reward_providers.rnd_reward_provider import (
     RNDRewardProvider,
 )
+from mlagents.trainers.torch.components.reward_providers.diverse_reward_provider import (
+    DiverseRewardProvider,
+)
 
 from mlagents_envs.base_env import BehaviorSpec
 
     RewardSignalType.GAIL: GAILRewardProvider,
     RewardSignalType.RND: RNDRewardProvider,
+    RewardSignalType.DIVERSE: DiverseRewardProvider,
 }
 
 

ml-agents/mlagents/trainers/sac/optimizer_torch.py

 
 logger = get_logger(__name__)
 
+from mlagents.trainers.torch.action_flattener import ActionFlattener
+from mlagents_envs.base_env import ObservationType
+from mlagents.trainers.torch.networks import NetworkBody
+from mlagents_envs.base_env import BehaviorSpec
+
+from mlagents.trainers.torch.layers import linear_layer, Initialization
+
+
+class DiverseNetworkVariational(torch.nn.Module):
+    EPSILON = 1e-10
+    STRENGTH =  0.1#1.0
+
+    # gradient_penalty_weight = 10.0
+    z_size = 128
+    alpha = 0.0005
+    mutual_information = 10000  # 0.5
+    EPSILON = 1e-7
+    initial_beta = 0.0
+
+    def __init__(self, specs: BehaviorSpec, settings) -> None:
+        super().__init__()
+        self._use_actions = True
+        sigma_start = 0.5
+        print(
+            "VARIATIONAL : Settings : strength:",
+            self.STRENGTH,
+            " use_actions:",
+            self._use_actions,
+            " mutual_information : ",
+            self.mutual_information,
+            "Sigma_Start : ",
+            sigma_start,
+        )
+        # state_encoder_settings = settings
+        state_encoder_settings = NetworkSettings(normalize=True, num_layers=1)
+        if state_encoder_settings.memory is not None:
+            state_encoder_settings.memory = None
+            logger.warning(
+                "memory was specified in network_settings but is not supported. It is being ignored."
+            )
+        self._action_flattener = ActionFlattener(specs.action_spec)
+        new_spec = [
+            spec
+            for spec in specs.observation_specs
+            if spec.observation_type != ObservationType.GOAL_SIGNAL
+        ]
+        diverse_spec = [
+            spec
+            for spec in specs.observation_specs
+            if spec.observation_type == ObservationType.GOAL_SIGNAL
+        ][0]
+
+        print(" > ", new_spec, "\n\n\n", " >> ", diverse_spec)
+        self._all_obs_specs = specs.observation_specs
+
+        self.diverse_size = diverse_spec.shape[0]
+
+        if self._use_actions:
+            self._encoder = NetworkBody(
+                new_spec, state_encoder_settings, self._action_flattener.flattened_size
+            )
+        else:
+            self._encoder = NetworkBody(new_spec, state_encoder_settings)
+
+        self._z_sigma = torch.nn.Parameter(
+            sigma_start * torch.ones((self.z_size), dtype=torch.float),
+            requires_grad=True,
+        )
+        # self._z_mu_layer = linear_layer(
+        #     state_encoder_settings.hidden_units,
+        #     self.z_size,
+        #     kernel_init=Initialization.KaimingHeNormal,
+        #     kernel_gain=0.1,
+        # )
+        self._beta = torch.nn.Parameter(
+            torch.tensor(self.initial_beta, dtype=torch.float), requires_grad=False
+        )
+
+        self._last_layer = torch.nn.Linear(self.z_size, self.diverse_size)
+        self._diverse_index = -1
+        self._max_index = len(specs.observation_specs)
+        for i, spec in enumerate(specs.observation_specs):
+            if spec.observation_type == ObservationType.GOAL_SIGNAL:
+                self._diverse_index = i
+
+    def predict(
+        self, obs_input, action_input, detach_action=False, var_noise=True
+    ) -> torch.Tensor:
+        # Convert to tensors
+        tensor_obs = [
+            obs
+            for obs, spec in zip(obs_input, self._all_obs_specs)
+            if spec.observation_type != ObservationType.GOAL_SIGNAL
+        ]
+        if self._use_actions:
+            action = self._action_flattener.forward(action_input).reshape(
+                -1, self._action_flattener.flattened_size
+            )
+            if detach_action:
+                action = action.detach()
+            hidden, _ = self._encoder.forward(tensor_obs, action)
+        else:
+            hidden, _ = self._encoder.forward(tensor_obs)
+
+        # add a VAE (like in VAIL ?)
+        # z_mu = self._z_mu_layer(hidden)
+        z_mu = hidden  # self._z_mu_layer(hidden)
+        hidden = torch.normal(z_mu, self._z_sigma * var_noise)
+
+        prediction = torch.softmax(self._last_layer(hidden), dim=1)
+        return prediction, z_mu
+
+    def copy_normalization(self, thing):
+        self._encoder.processors[0].copy_normalization(thing.processors[1])
+
+    def rewards(
+        self, obs_input, action_input, detach_action=False, var_noise=True
+    ) -> torch.Tensor:
+        truth = obs_input[self._diverse_index]
+        prediction, _ = self.predict(obs_input, action_input, detach_action, var_noise)
+        rewards = torch.log(
+            torch.sum((prediction * truth), dim=1) + self.EPSILON
+        )# - np.log(1 / self.diverse_size)  # Center around 0
+        return rewards
+
+    def loss(
+        self, obs_input, action_input, masks, detach_action=True, var_noise=True
+    ) -> torch.Tensor:
+        # print( ">>> ",obs_input[self._diverse_index][0],self.predict(obs_input, action_input, detach_action)[0], self.predict([x*0 for x in obs_input], action_input, detach_action * 0)[0] )
+        base_loss = -ModelUtils.masked_mean(
+            self.rewards(obs_input, action_input, detach_action, var_noise), masks
+        )
+
+        _, mu = self.predict(obs_input, action_input, detach_action, var_noise)
+        kl_loss = ModelUtils.masked_mean(
+            -torch.sum(
+                1 + (self._z_sigma ** 2).log() - 0.5 * mu ** 2
+                # - 0.5 * mu_expert ** 2
+                - (self._z_sigma ** 2),
+                dim=1,
+            ),
+            masks,
+        )
+        vail_loss = self._beta * (kl_loss - self.mutual_information)
+        with torch.no_grad():
+            self._beta.data = torch.max(
+                self._beta + self.alpha * (kl_loss - self.mutual_information),
+                torch.tensor(0.0),
+            )
+        total_loss = base_loss + vail_loss
+
+        return total_loss, base_loss, kl_loss, vail_loss, self._beta
+
+
+class DiverseNetwork(torch.nn.Module):
+    EPSILON = 1e-10
+    STRENGTH = 1.0
+
+    def __init__(self, specs: BehaviorSpec, settings) -> None:
+        super().__init__()
+        self._use_actions = True
+        print("Settings : strength:", self.STRENGTH, " use_actions:", self._use_actions)
+        # state_encoder_settings = settings
+        state_encoder_settings = NetworkSettings(True)
+        if state_encoder_settings.memory is not None:
+            state_encoder_settings.memory = None
+            logger.warning(
+                "memory was specified in network_settings but is not supported. It is being ignored."
+            )
+        self._action_flattener = ActionFlattener(specs.action_spec)
+        new_spec = [
+            spec
+            for spec in specs.observation_specs
+            if spec.observation_type != ObservationType.GOAL_SIGNAL
+        ]
+        diverse_spec = [
+            spec
+            for spec in specs.observation_specs
+            if spec.observation_type == ObservationType.GOAL_SIGNAL
+        ][0]
+
+        print(" > ", new_spec, "\n\n\n", " >> ", diverse_spec)
+        self._all_obs_specs = specs.observation_specs
+
+        self.diverse_size = diverse_spec.shape[0]
+
+        if self._use_actions:
+            self._encoder = NetworkBody(
+                new_spec, state_encoder_settings, self._action_flattener.flattened_size
+            )
+        else:
+            self._encoder = NetworkBody(new_spec, state_encoder_settings)
+        self._last_layer = torch.nn.Linear(
+            state_encoder_settings.hidden_units, self.diverse_size
+        )
+        self._diverse_index = -1
+        self._max_index = len(specs.observation_specs)
+        for i, spec in enumerate(specs.observation_specs):
+            if spec.observation_type == ObservationType.GOAL_SIGNAL:
+                self._diverse_index = i
+
+    def predict(self, obs_input, action_input, detach_action=False) -> torch.Tensor:
+        # Convert to tensors
+        tensor_obs = [
+            obs
+            for obs, spec in zip(obs_input, self._all_obs_specs)
+            if spec.observation_type != ObservationType.GOAL_SIGNAL
+        ]
+        if self._use_actions:
+            action = self._action_flattener.forward(action_input).reshape(
+                -1, self._action_flattener.flattened_size
+            )
+            if detach_action:
+                action = action.detach()
+            hidden, _ = self._encoder.forward(tensor_obs, action)
+        else:
+            hidden, _ = self._encoder.forward(tensor_obs)
+
+        # add a VAE (like in VAIL ?)
+
+        prediction = torch.softmax(self._last_layer(hidden), dim=1)
+        return prediction
+
+    def copy_normalization(self, thing):
+        self._encoder.processors[0].copy_normalization(thing.processors[1])
+
+    def rewards(
+        self, obs_input, action_input, detach_action=False, var_noise=False
+    ) -> torch.Tensor:
+        truth = obs_input[self._diverse_index]
+        prediction = self.predict(obs_input, action_input, detach_action)
+        rewards = torch.log(torch.sum((prediction * truth), dim=1) + self.EPSILON)
+        return rewards
+
+    def loss(self, obs_input, action_input, masks, detach_action=True) -> torch.Tensor:
+        # print( ">>> ",obs_input[self._diverse_index][0],self.predict(obs_input, action_input, detach_action)[0], self.predict([x*0 for x in obs_input], action_input, detach_action * 0)[0] )
+        return -ModelUtils.masked_mean(
+            self.rewards(obs_input, action_input, detach_action), masks
+        )
+
 
 class TorchSACOptimizer(TorchOptimizer):
     class PolicyValueNetwork(nn.Module):
             self._critic.parameters()
         )
 
+        # self._mede_network = DiverseNetwork(
+        self._mede_network = DiverseNetworkVariational(
+            self.policy.behavior_spec, self.policy.network_settings
+        )
+        self._mede_optimizer = torch.optim.Adam(
+            list(self._mede_network.parameters()), lr=hyperparameters.learning_rate
+        )
+
         logger.debug("value_vars")
         for param in value_params:
             logger.debug(param.shape)
         q1p_out: Dict[str, torch.Tensor],
         q2p_out: Dict[str, torch.Tensor],
         loss_masks: torch.Tensor,
+        obs,
+        act,
     ) -> torch.Tensor:
         min_policy_qs = {}
         with torch.no_grad():
         if self._action_spec.discrete_size <= 0:
             for name in values.keys():
                 with torch.no_grad():
-                    v_backup = min_policy_qs[name] - torch.sum(
-                        _cont_ent_coef * log_probs.continuous_tensor, dim=1
+                    v_backup = (
+                        min_policy_qs[name]
+                        - _cont_ent_coef * torch.sum( log_probs.continuous_tensor, dim=1)
+                        + self._mede_network.STRENGTH
+                        * self._mede_network.rewards(obs, act, var_noise=False)
                     )
                 value_loss = 0.5 * ModelUtils.masked_mean(
                     torch.nn.functional.mse_loss(values[name], v_backup), loss_masks
             )
             for name in values.keys():
                 with torch.no_grad():
-                    v_backup = min_policy_qs[name] - torch.mean(
-                        branched_ent_bonus, axis=0
+                    v_backup = (
+                        min_policy_qs[name]
+                        - torch.mean(branched_ent_bonus, axis=0)
+                        + self._mede_network.STRENGTH
+                        * self._mede_network.rewards(obs, act)
+                    print("The discrete case is much more complicated than that")
                     # Add continuous entropy bonus to minimum Q
                     if self._action_spec.continuous_size > 0:
                         v_backup += torch.sum(
         log_probs: ActionLogProbs,
         q1p_outs: Dict[str, torch.Tensor],
         loss_masks: torch.Tensor,
+        obs,
+        act,
     ) -> torch.Tensor:
         _cont_ent_coef, _disc_ent_coef = (
             self._log_ent_coef.continuous,
             all_mean_q1 = mean_q1
         if self._action_spec.continuous_size > 0:
             cont_log_probs = log_probs.continuous_tensor
-            batch_policy_loss += torch.mean(
-                _cont_ent_coef * cont_log_probs - all_mean_q1.unsqueeze(1), dim=1
-            )
+            batch_policy_loss += _cont_ent_coef * torch.sum(cont_log_probs, dim=1) - all_mean_q1.unsqueeze(1)
+        batch_policy_loss += -self._mede_network.STRENGTH * self._mede_network.rewards(
+            obs, act, var_noise=False
+        )
         policy_loss = ModelUtils.masked_mean(batch_policy_loss, loss_masks)
 
         return policy_loss
         if self._action_spec.continuous_size > 0:
             with torch.no_grad():
                 cont_log_probs = log_probs.continuous_tensor
-                target_current_diff = torch.sum(
-                    cont_log_probs + self.target_entropy.continuous, dim=1
+                target_current_diff = (
+                    torch.sum(cont_log_probs, dim=1)
+                    + self.target_entropy.continuous
+                # print(self.target_entropy.continuous, cont_log_probs, torch.sum(
+                #     cont_log_probs, dim=1) + self.target_entropy.continuous)
             # We update all the _cont_ent_coef as one block
             entropy_loss += -1 * ModelUtils.masked_mean(
                 _cont_ent_coef * target_current_diff, loss_masks
         self.target_network.network_body.copy_normalization(
             self.policy.actor.network_body
         )
+        self._mede_network.copy_normalization(self.policy.actor.network_body)
         self._critic.network_body.copy_normalization(self.policy.actor.network_body)
         sampled_actions, log_probs, _, _, = self.policy.actor.get_action_and_stats(
             current_obs,
             q1_stream, q2_stream, target_values, dones, rewards, masks
         )
         value_loss = self.sac_value_loss(
-            log_probs, value_estimates, q1p_out, q2p_out, masks
+            log_probs,
+            value_estimates,
+            q1p_out,
+            q2p_out,
+            masks,
+            current_obs,
+            sampled_actions,
-        policy_loss = self.sac_policy_loss(log_probs, q1p_out, masks)
+        policy_loss = self.sac_policy_loss(
+            log_probs, q1p_out, masks, current_obs, sampled_actions
+        )
         entropy_loss = self.sac_entropy_loss(log_probs, masks)
 
         total_value_loss = q1_loss + q2_loss
         entropy_loss.backward()
         self.entropy_optimizer.step()
 
+        mede_loss, base_loss, kl_loss, vail_loss, beta = self._mede_network.loss(
+            current_obs, sampled_actions, masks
+        )
+        # mede_loss = self._mede_network.loss(current_obs, sampled_actions, masks)
+        ModelUtils.update_learning_rate(self._mede_optimizer, decay_lr)
+        self._mede_optimizer.zero_grad()
+        mede_loss.backward()
+        self._mede_optimizer.step()
+
         # Update target network
         ModelUtils.soft_update(self._critic, self.target_network, self.tau)
         update_stats = {
                 torch.exp(self._log_ent_coef.continuous)
             ).item(),
             "Policy/Learning Rate": decay_lr,
+            "Policy/Entropy Loss": entropy_loss.item(),
+            "Policy/MEDE Loss": mede_loss.item(),
+            "Policy/MEDE Base": base_loss.item(),
+            "Policy/MEDE Variational": vail_loss.item(),
+            "Policy/MEDE KL": kl_loss.item(),
+            "Policy/MEDE beta": beta.item(),
         }
 
         return update_stats
             "Optimizer:policy_optimizer": self.policy_optimizer,
             "Optimizer:value_optimizer": self.value_optimizer,
             "Optimizer:entropy_optimizer": self.entropy_optimizer,
+            "Optimizer:mede_optimizer": self._mede_optimizer,
+            "Optimizer:mede_network": self._mede_network,
         }
         for reward_provider in self.reward_signals.values():
             modules.update(reward_provider.get_modules())

Project/Assets/ML-Agents/Examples/Walker/TFModels/Walker-10M.onnx.meta

+fileFormatVersion: 2
+guid: c4c1c9de2772f48e8b0cc2cdd62ce8c5
+ScriptedImporter:
+  internalIDToNameTable: []
+  externalObjects: {}
+  serializedVersion: 2
+  userData: 
+  assetBundleName: 
+  assetBundleVariant: 
+  script: {fileID: 11500000, guid: 683b6cb6d0a474744822c888b46772c9, type: 3}
+  optimizeModel: 1
+  forceArbitraryBatchSize: 1
+  treatErrorsAsWarnings: 0
+  importMode: 1

Project/Assets/ML-Agents/Examples/Walker/TFModels/Walker-diverse-no-extrinsic.onnx.meta

+fileFormatVersion: 2
+guid: e97d22662d00b43ed999531846572c36
+ScriptedImporter:
+  internalIDToNameTable: []
+  externalObjects: {}
+  serializedVersion: 2
+  userData: 
+  assetBundleName: 
+  assetBundleVariant: 
+  script: {fileID: 11500000, guid: 683b6cb6d0a474744822c888b46772c9, type: 3}
+  optimizeModel: 1
+  forceArbitraryBatchSize: 1
+  treatErrorsAsWarnings: 0
+  importMode: 1

Project/Assets/ML-Agents/Examples/Walker/TFModels/Walker-diverse-r02-bigger.onnx.meta

+fileFormatVersion: 2
+guid: 7c800225c636c4e299f1decdfa9b9029
+ScriptedImporter:
+  internalIDToNameTable: []
+  externalObjects: {}
+  serializedVersion: 2
+  userData: 
+  assetBundleName: 
+  assetBundleVariant: 
+  script: {fileID: 11500000, guid: 683b6cb6d0a474744822c888b46772c9, type: 3}
+  optimizeModel: 1
+  forceArbitraryBatchSize: 1
+  treatErrorsAsWarnings: 0
+  importMode: 1

Project/Assets/ML-Agents/Examples/Walker/TFModels/Walker-diverse-r05-bigger.onnx.meta

+fileFormatVersion: 2
+guid: 55f976c29f6ff4d2bbe6dff72645410a
+ScriptedImporter:
+  internalIDToNameTable: []
+  externalObjects: {}
+  serializedVersion: 2
+  userData: 
+  assetBundleName: 
+  assetBundleVariant: 
+  script: {fileID: 11500000, guid: 683b6cb6d0a474744822c888b46772c9, type: 3}
+  optimizeModel: 1
+  forceArbitraryBatchSize: 1
+  treatErrorsAsWarnings: 0
+  importMode: 1

Project/Assets/ML-Agents/Examples/Walker/TFModels/Walker-extrinsic-log-diverse.onnx.meta

+fileFormatVersion: 2
+guid: a9ceaa0e80ae34d229e1a2277d3388e5
+ScriptedImporter:
+  internalIDToNameTable: []
+  externalObjects: {}
+  serializedVersion: 2
+  userData: 
+  assetBundleName: 
+  assetBundleVariant: 
+  script: {fileID: 11500000, guid: 683b6cb6d0a474744822c888b46772c9, type: 3}
+  optimizeModel: 1
+  forceArbitraryBatchSize: 1
+  treatErrorsAsWarnings: 0
+  importMode: 1

Project/Assets/ML-Agents/Examples/Walker/TFModels/Walker-new-reward-1.onnx.meta

+fileFormatVersion: 2
+guid: 29312cd291f9d4f2d91c7272a2a14324
+ScriptedImporter:
+  internalIDToNameTable: []
+  externalObjects: {}
+  serializedVersion: 2
+  userData: 
+  assetBundleName: 
+  assetBundleVariant: 
+  script: {fileID: 11500000, guid: 683b6cb6d0a474744822c888b46772c9, type: 3}
+  optimizeModel: 1
+  forceArbitraryBatchSize: 1
+  treatErrorsAsWarnings: 0
+  importMode: 1

Project/Assets/ML-Agents/Examples/Walker/TFModels/mede-walker-crazy-mutual-10000-nogound-penalty.onnx.meta

+fileFormatVersion: 2
+guid: 7ce5ceea4d8ff4d4a837a510deed0b0e
+ScriptedImporter:
+  internalIDToNameTable: []
+  externalObjects: {}
+  serializedVersion: 2
+  userData: 
+  assetBundleName: 
+  assetBundleVariant: 
+  script: {fileID: 11500000, guid: 683b6cb6d0a474744822c888b46772c9, type: 3}
+  optimizeModel: 1
+  forceArbitraryBatchSize: 1
+  treatErrorsAsWarnings: 0
+  importMode: 1

Project/Assets/ML-Agents/Examples/Walker/TFModels/mede-walker-crazy-mutual-10000-nogound-penalty-01strength.onnx.meta

+fileFormatVersion: 2
+guid: 371ea8d3fe4bf4cde9da72b71d914c52
+ScriptedImporter:
+  internalIDToNameTable: []
+  externalObjects: {}
+  serializedVersion: 2
+  userData: 
+  assetBundleName: 
+  assetBundleVariant: 
+  script: {fileID: 11500000, guid: 683b6cb6d0a474744822c888b46772c9, type: 3}
+  optimizeModel: 1
+  forceArbitraryBatchSize: 1
+  treatErrorsAsWarnings: 0
+  importMode: 1

ml-agents/mlagents/trainers/torch/components/reward_providers/diverse_reward_provider.py

+import numpy as np
+from typing import Dict
+from mlagents.torch_utils import torch
+
+from mlagents_envs.base_env import ObservationType
+from mlagents.trainers.buffer import AgentBuffer
+from mlagents.trainers.torch.components.reward_providers.base_reward_provider import (
+    BaseRewardProvider,
+)
+from mlagents.trainers.settings import DiverseSettings
+from mlagents.trainers.torch.action_flattener import ActionFlattener
+from mlagents.trainers.torch.agent_action import AgentAction
+
+from mlagents_envs.base_env import BehaviorSpec
+from mlagents_envs import logging_util
+from mlagents.trainers.torch.utils import ModelUtils
+from mlagents.trainers.torch.networks import NetworkBody
+from mlagents.trainers.trajectory import ObsUtil
+
+logger = logging_util.get_logger(__name__)
+
+
+class DiverseRewardProvider(BaseRewardProvider):
+    # From https://arxiv.org/pdf/1802.06070.pdf
+    def __init__(self, specs: BehaviorSpec, settings: DiverseSettings) -> None:
+        super().__init__(specs, settings)
+        self._ignore_done = False  # Tried with false. Bias for staying alive.
+        self._use_actions = False
+
+        self._network = DiverseNetwork(specs, settings, self._use_actions)
+        self.optimizer = torch.optim.SGD(
+            self._network.parameters(), lr=settings.learning_rate
+        )
+        self._diverse_index = -1
+        self._max_index = len(specs.observation_specs)
+        for i, spec in enumerate(specs.observation_specs):
+            if spec.observation_type == ObservationType.GOAL_SIGNAL:
+                self._diverse_index = i
+
+    def evaluate(self, mini_batch: AgentBuffer) -> np.ndarray:
+        with torch.no_grad():
+            prediction = self._network(mini_batch)
+            truth = ModelUtils.list_to_tensor(
+                ObsUtil.from_buffer(mini_batch, self._max_index)[self._diverse_index]
+            )
+            # print(prediction[0,:], truth[0,:], torch.log(torch.sum((prediction * truth), dim=1) + 1e-10)[0], (torch.log(torch.sum((prediction * truth), dim=1))- np.log(1 / self._network.diverse_size))[0])
+            rewards = torch.log(
+                torch.sum((prediction * truth), dim=1) + 1e-10
+            ) - np.log(1 / self._network.diverse_size)
+
+        return rewards.detach().cpu().numpy()
+
+    def update(self, mini_batch: AgentBuffer) -> Dict[str, np.ndarray]:
+        all_loss = 0
+        for _ in range(1):
+            prediction = self._network(mini_batch)
+            truth = ModelUtils.list_to_tensor(
+                ObsUtil.from_buffer(mini_batch, self._max_index)[self._diverse_index]
+            )
+            # loss = torch.mean(
+            #     torch.sum(-torch.log(prediction + 1e-10) * truth, dim=1), dim=0
+            # )
+            loss = -torch.mean(torch.log(torch.sum((prediction * truth), dim=1)))
+            self.optimizer.zero_grad()
+            loss.backward()
+            self.optimizer.step()
+            all_loss += loss.item()
+        return {"Losses/DIVERSE Loss": all_loss}
+
+    def get_modules(self):
+        return {f"Module:{self.name}": self._network}
+
+
+class DiverseNetwork(torch.nn.Module):
+    EPSILON = 1e-10
+
+    def __init__(
+        self, specs: BehaviorSpec, settings: DiverseSettings, use_actions: bool
+    ) -> None:
+        super().__init__()
+        self._use_actions = use_actions
+        state_encoder_settings = settings.network_settings
+        if state_encoder_settings.memory is not None:
+            state_encoder_settings.memory = None
+            logger.warning(
+                "memory was specified in network_settings but is not supported. It is being ignored."
+            )
+        self._action_flattener = ActionFlattener(specs.action_spec)
+        new_spec = [
+            spec
+            for spec in specs.observation_specs
+            if spec.observation_type != ObservationType.GOAL_SIGNAL
+        ]
+        diverse_spec = [
+            spec
+            for spec in specs.observation_specs
+            if spec.observation_type == ObservationType.GOAL_SIGNAL
+        ][0]
+
+        print(" > ", new_spec, "\n\n\n", " >> ", diverse_spec)
+        self._all_obs_specs = specs.observation_specs
+
+        self.diverse_size = diverse_spec.shape[0]
+
+        if self._use_actions:
+            self._encoder = NetworkBody(
+                new_spec, state_encoder_settings, self._action_flattener.flattened_size
+            )
+        else:
+            self._encoder = NetworkBody(new_spec, state_encoder_settings)
+        self._last_layer = torch.nn.Linear(
+            state_encoder_settings.hidden_units, self.diverse_size
+        )
+
+    def forward(self, mini_batch: AgentBuffer) -> torch.Tensor:
+        n_obs = len(self._encoder.processors) + 1
+        np_obs = ObsUtil.from_buffer_next(mini_batch, n_obs)
+        # Convert to tensors
+        tensor_obs = [
+            ModelUtils.list_to_tensor(obs)
+            for obs, spec in zip(np_obs, self._all_obs_specs)
+            if spec.observation_type != ObservationType.GOAL_SIGNAL
+        ]
+
+        if self._use_actions:
+            action = self._action_flattener.forward(AgentAction.from_buffer(mini_batch))
+            hidden, _ = self._encoder.forward(tensor_obs, action)
+        else:
+            hidden, _ = self._encoder.forward(tensor_obs)
+        self._encoder.update_normalization(mini_batch)
+
+        prediction = torch.softmax(self._last_layer(hidden), dim=1)
+        return prediction