Open source code for adaptive behavior policy sharing(ABPS) paper.

PiperOrigin-RevId: 288816072
rally12 · Jan 9, 2020 · 0095e70 · 0095e70
1 parent 2033584
commit 0095e70
Show file tree

Hide file tree

Showing 16 changed files with 6,207 additions and 0 deletions.
diff --git a/.travis.yml b/.travis.yml
@@ -8,6 +8,7 @@ python:
   - "3.6"
 env:
   matrix:
+    - PROJECT="abps"
     - PROJECT="albert"
     - PROJECT="attribution"
     - PROJECT="axial"

diff --git a/abps/README.md b/abps/README.md
@@ -0,0 +1,3 @@
+# Adaptive Behavior Policy Sharing for data-efficient RL training
+This package includes the code used to run the experiments in the NeurIPS 2019 Deep RL workshop/Optimization foundation for RL workshop paper [Data Efficient Training for Reinforcement Learning with Adaptive Behavior Policy Sharing](https://drive.google.com/file/d/1bCJg5tQQAq1CCo8IXfXjcYQk1uQjV0bp/view).
+
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,3 @@
		# Adaptive Behavior Policy Sharing for data-efficient RL training
		This package includes the code used to run the experiments in the NeurIPS 2019 Deep RL workshop/Optimization foundation for RL workshop paper [Data Efficient Training for Reinforcement Learning with Adaptive Behavior Policy Sharing](https://drive.google.com/file/d/1bCJg5tQQAq1CCo8IXfXjcYQk1uQjV0bp/view).