About

This Repository contains an implementation of ICM( Intrinsic Curiosity reward ) module by plugging it into three different algorithms i.e. DQN, A3C, and PPO. This whole code is based on the ICM paper introduced by Deepak Pathak, Pulkit Agrawal, Alexei A. Efros, and Trevor Darrell in the paper Curiosity-driven Exploration by Self-supervised Prediction.

Organization

The repository is divided into 3 parts corresponding to each algorithm implementation. PPO and A3C's implementation is a group of Python scripts in a directory while DQN's implementation is in a Jupyter notebook. For running clone the repository and execute the code normally.

Demo for PPO and ICM:

1. Google Drive: Demo

References:

1. A3C + ICM: https://youtu.be/O5Z-3q-J18I?si=bg4ytpGm63BtvltE
2. PPO + ICM: https://github.com/adik993/ppo-pytorch
3. https://pathak22.github.io/noreward-rl/