Skip to content

rlenvs.md

Latest commit

 

History

History
61 lines (54 loc) · 6.06 KB

rlenvs.md

File metadata and controls

61 lines (54 loc) · 6.06 KB

ReinforcementLearningEnvironments.jl

Built-in Environments

<table>
<th colspan="2">Traits</th><th> 1 </th><th> 2 </th><th> 3 </th><th> 4 </th><th> 5 </th><th> 6 </th><th> 7 </th><th> 8 </th><th> 9 </th><th> 10 </th><th> 11 </th><th> 12 </th><th> 13 </th><tr> <th rowspan="2"> ActionStyle </th><th> MinimalActionSet </th><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> </td> <td> ✔ </td><td> </td> <td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td></tr>
<tr> <th> FullActionSet </th><td> </td> <td> </td> <td> </td> <td> ✔ </td><td> </td> <td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> </tr>
<tr> <th rowspan="3"> ChanceStyle </th><th> Stochastic </th><td> ✔ </td><td> </td> <td> ✔ </td><td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td></tr>
<tr> <th> Deterministic </th><td> </td> <td> ✔ </td><td> </td> <td> </td> <td> ✔ </td><td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> </tr>
<tr> <th> ExplicitStochastic </th><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> ✔ </td><td> ✔ </td><td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> </tr>
<tr> <th rowspan="2"> DefaultStateStyle </th><th> Observation </th><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> </td> <td> ✔ </td><td> </td> <td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td></tr>
<tr> <th> InformationSet </th><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> ✔ </td><td> </td> <td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> </tr>
<tr> <th rowspan="2"> DynamicStyle </th><th> Simultaneous </th><td> </td> <td> </td> <td> </td> <td> </td> <td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> </tr>
<tr> <th> Sequential </th><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> </td> <td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td></tr>
<tr> <th rowspan="2"> InformationStyle </th><th> PerfectInformation </th><td> </td> <td> ✔ </td><td> </td> <td> </td> <td> </td> <td> ✔ </td><td> </td> <td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> </tr>
<tr> <th> ImperfectInformation </th><td> ✔ </td><td> </td> <td> ✔ </td><td> ✔ </td><td> ✔ </td><td> </td> <td> ✔ </td><td> </td> <td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td></tr>
<tr> <th rowspan="2"> NumAgentStyle </th><th> MultiAgent </th><td> </td> <td> </td> <td> </td> <td> </td> <td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> </tr>
<tr> <th> SingleAgent </th><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td></tr>
<tr> <th rowspan="2"> RewardStyle </th><th> TerminalReward </th><td> ✔ </td><td> ✔ </td><td> </td> <td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> </tr>
<tr> <th> StepReward </th><td> </td> <td> </td> <td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td></tr>
<tr> <th rowspan="3"> StateStyle </th><th> Observation </th><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> </td> <td> ✔ </td><td> </td> <td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td></tr>
<tr> <th> InformationSet </th><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> ✔ </td><td> </td> <td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> </tr>
<tr> <th> InternalState </th><td> </td> <td> </td> <td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> </tr>
<tr> <th rowspan="4"> UtilityStyle </th><th> GeneralSum </th><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> ✔ </td><td> ✔ </td><td> ✔ </td><td> ✔ </td></tr>
<tr> <th> ZeroSum </th><td> </td> <td> </td> <td> </td> <td> </td> <td> ✔ </td><td> ✔ </td><td> </td> <td> </td> <td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> </tr>
<tr> <th> ConstantSum </th><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> </tr>
<tr> <th> IdenticalUtility </th><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> ✔ </td><td> </td> <td> </td> <td> </td> <td> </td> <td> </td> <td> </td> </tr>
</table>
<ol><li> MultiArmBanditsEnv </li>
<li> RandomWalk1D </li>
<li> TigerProblemEnv </li>
<li> MontyHallEnv </li>
<li> RockPaperScissorsEnv </li>
<li> TicTacToeEnv </li>
<li> TinyHanabiEnv </li>
<li> PigEnv </li>
<li> KuhnPokerEnv </li>
<li> AcrobotEnv </li>
<li> CartPoleEnv </li>
<li> MountainCarEnv </li>
<li> PendulumEnv </li>
</ol>

Note: Many traits are borrowed from OpenSpiel.

3-rd Party Environments

Environment Name Dependent Package Name Description
AtariEnv ArcadeLearningEnvironment.jl
GymEnv PyCall.jl
OpenSpielEnv OpenSpiel.jl
SnakeGameEnv SnakeGames.jl SingleAgent/Multi-Agent, FullActionSet/MinimalActionSet
#list-of-environments GridWorlds.jl Environments in this package support the interfaces defined in RLBase 
Modules = [ReinforcementLearningEnvironments]