/TD3-and-Extensions

PyTorch implementation of Twin Delayed Deep Deterministic Policy Gradient (TD3) - including additional Extension to improve the algorithm's performance.

Primary LanguagePython

TD3_and_Extensions

Setup

Run

Pendulum-v0: Pendulum-nstep-PER

LunarLanderContinuous-v2

Results

TODOs:

  • add per [X] /ere
  • add n-step [X]
  • add munchausenRL
  • add D2RL
  • Make test runs for BulletEnvs
  • Make comparison to SAC

Author