DistributedRL-Pytorch-Ray icon indicating copy to clipboard operation
DistributedRL-Pytorch-Ray copied to clipboard

Published 20 hours ago • verified publisher icon seolhokim
→

Metadata

Distributed RL Implementation using Pytorch and Ray (ApeX(Ape-X), A3C, Distributed-PPO(DPPO), Impala)

DistributedRL-Pytorch-Ray

Algorithm

  • A3C
  • DPPO
  • Ape-X
    • (Discrete version)
  • Impala

Tested Environment

Continuous

  • MountainCarContinuous-v0
  • Mujoco Benchmarks(Hopper,... etc)

Discrete

  • CartPole-v1
  • LunarLander-v2

TODO

Fix

  • Fix cuda environment clock time
  • Update Impala multi learner version
  • Check Ape-X performance
    • Performance does not go up in the middle.
  • Experiment distributed environment.
    • Implemented to use only one computer.

Add

  • add LASER
  • add R2D2
  • add NGU
  • add Agent57
  • test more environments

← Metadata

27
Stars
3
Forks
Watchers

Owner

verified publisher icon seolhokim

Metadata

Distributed RL Implementation using Pytorch and Ray (ApeX(Ape-X), A3C, Distributed-PPO(DPPO), Impala)

Back