Comparison of Proximal Policy Optimization(PPO), Soft Actor Critic(SAC) and Twin Delayed(TD3) in Solving Bipedal Walker Task + Genetic Algorithm Hyperparameter Optimization

Requirements

Code environment is managed via Anaconda in this project.

To create an environment and install all dependencies:

conda env create -f environment.yml

To clone and run this project locally:

git clone https://github.com/ChengyuanSha/RL-Bipedal-Walker

File Structure

• PPO folder by Chengyuan Sha:
  • PPO.ipynb
• SAC folder by Dongjun Jin:
  • main.ipynb
• TD3 folder by Jinting Zhang:
  • main.ipynb