hiro/ray
1
0
Fork 0
mirror of https://github.com/vale981/ray synced 2025-03-06 10:31:39 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions

[hotfix] Fix table formatting (#10687)

Browse source
This commit is contained in:
Eric Liang 2020-09-09 16:08:54 -07:00 • committed by GitHub
parent 8a1caf6279
commit f7d5aa46a3
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 4 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

7
doc/source/rllib-algorithms.rst
View file

@ -42,7 +42,7 @@ Algorithm Frameworks Discrete Actions Continuous Acti
----------------------------- ---------------------------------------------------------------------------------------
`Fully Independent Learning`_ Depends on bootstrapped algorithm
----------------------------- ---------------------------------------------------------------------------------------
`Shared Critic Methods`_ Depends on bootstrapped algorithm
`Shared Critic Methods`_ Depends on bootstrapped algorithm
============================= =======================================================================================
@ -734,5 +734,6 @@ Tuned examples: `waterworld <https://github.com/ray-project/ray/blob/master/rlli
Shared Critic Methods
--------------------------
`[instructions] <https://docs.ray.io/en/master/rllib-env.html#implementing-a-centralized-critic>`__Shared critic methods are when all agents use a single parameter shared critic network (in some cases with access to more of the observation space than agents can see). Note that many specialized multi-agent algorithms such as MADDPG are mostly shared critic forms of their single-agent algorithm (DDPG in the case of MADDPG).
Tuned examples: `TwoStepGame <https://github.com/ray-project/ray/blob/master/rllib/examples/centralized_critic_2.py>`__
`[instructions] <https://docs.ray.io/en/master/rllib-env.html#implementing-a-centralized-critic>`__ Shared critic methods are when all agents use a single parameter shared critic network (in some cases with access to more of the observation space than agents can see). Note that many specialized multi-agent algorithms such as MADDPG are mostly shared critic forms of their single-agent algorithm (DDPG in the case of MADDPG).
Tuned examples: `TwoStepGame <https://github.com/ray-project/ray/blob/master/rllib/examples/centralized_critic_2.py>`__