Skip to content

Latest commit

 

History

History
10 lines (9 loc) · 411 Bytes

README.md

File metadata and controls

10 lines (9 loc) · 411 Bytes
Raw

DelayResolvedRL

Implementation of the DRDQN algrotihm described in our CIKM'21 paper titled "Revisiting State Augmentation methods for Reinforcement Learning with Stochastic Delays"

System requirements

  • Python 3.6.9
  • Tensorflow 2.1
  • OpenAI Gym 0.18.0
The repository includes experiemnts on standard Gym enviroments (and W-maze) with both constant and stochastic delays.