/curiosity-driven-exploration-pytorch

Curiosity-driven Exploration by Self-supervised Prediction

Primary LanguagePythonMIT LicenseMIT

Curiosity-driven Exploration by Self-supervised Prediction

  • Advantage Actor critic [1]
  • Parallel Advantage Actor critic [2]
  • Curiosity-driven Exploration by Self-supervised Prediction [3] [5]
  • Proximal Policy Optimization Algorithms [4]

1. Setup

Requirements


2. How to Train

Modify the parameters in config.conf as you like.

python train.py

3. How to Eval

python eval.py

4. Loss/Reward Graph

  • Breakout Env image

References

[1] Actor-Critic Algorithms
[2] Efficient Parallel Methods for Deep Reinforcement Learning
[3] Curiosity-driven Exploration by Self-supervised Prediction
[4] Proximal Policy Optimization Algorithms
[5] Large-Scale Study of Curiosity-Driven Learning