wassname/ray
1
0
Fork 0
mirror of https://github.com/wassname/ray.git synced 2026-08-18 12:20:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

[RLlib] PyTorch version of ES (Evolution Strategies). (#8104)

Browse Source 
PyTorch version of Evolution Strategies (ES) Algo.
This commit is contained in:
Sven Mika
2020-04-20 21:47:28 +02:00
committed by GitHub
parent 9f3e9e7e9f
commit 3812bfedda
17 changed files with 276 additions and 121 deletions
Download Patch File Download Diff File Expand all files Collapse all files
doc/source/rllib-algorithms.rst
+2 -2
View File
@@ -13,7 +13,7 @@ Algorithm Frameworks Discrete Actions Continuous Actions Multi-
=================== ========== ======================= ================== =========== =====================
`A2C, A3C`_ tf + torch **Yes** `+parametric`_ **Yes** **Yes** `+RNN`_, `+autoreg`_
`ARS`_ tf **Yes** **Yes** No
`ES`_ tf **Yes** **Yes** No
`ES`_ tf + torch **Yes** **Yes** No
`DDPG`_, `TD3`_ tf + torch No **Yes** **Yes**
`APEX-DDPG`_ tf No **Yes** **Yes**
`DQN`_, `Rainbow`_ tf + torch **Yes** `+parametric`_ No **Yes**
@@ -422,7 +422,7 @@ Tuned examples: `CartPole-v0 <https://github.com/ray-project/ray/blob/master/rll
Evolution Strategies
--------------------
|tensorflow|
|pytorch| |tensorflow|
`[paper] <https://arxiv.org/abs/1703.03864>`__ `[implementation] <https://github.com/ray-project/ray/blob/master/rllib/agents/es/es.py>`__
Code here is adapted from https://github.com/openai/evolution-strategies-starter to execute in the distributed setting with Ray.
doc/source/rllib-toc.rst
+4 -4
View File
@@ -101,7 +101,7 @@ Algorithms
- |pytorch| |tensorflow| :ref:`Advantage Actor-Critic (A2C, A3C) <a3c>`
- |tensorflow| :ref:`Deep Deterministic Policy Gradients (DDPG, TD3) <ddpg>`
- |pytorch| |tensorflow| :ref:`Deep Deterministic Policy Gradients (DDPG, TD3) <ddpg>`
- |pytorch| |tensorflow| :ref:`Deep Q Networks (DQN, Rainbow, Parametric DQN) <dqn>`
@@ -109,13 +109,13 @@ Algorithms
- |pytorch| |tensorflow| :ref:`Proximal Policy Optimization (PPO) <ppo>`
- |tensorflow| :ref:`Soft Actor Critic (SAC) <sac>`
- |pytorch| |tensorflow| :ref:`Soft Actor Critic (SAC) <sac>`
* Derivative-free
- |tensorflow| :ref:`Augmented Random Search (ARS) <ars>`
- |tensorflow| :ref:`Evolution Strategies <es>`
- |pytorch| |tensorflow| :ref:`Evolution Strategies <es>`
* Multi-agent specific
@@ -124,7 +124,7 @@ Algorithms
* Offline
- |tensorflow| :ref:`Advantage Re-Weighted Imitation Learning (MARWIL) <marwil>`
- |pytorch| |tensorflow| :ref:`Advantage Re-Weighted Imitation Learning (MARWIL) <marwil>`
* Contextual bandits