/CDA-MBPO

Primary LanguagePython

CDA-MBPO: Corrected Data Aggregation for Model-based Policy Optimization

This repository contains a TensorFlow implementation of [CDA-MBPO]

Requirements

  1. Install MuJoCo 1.50 at ~/.mujoco/mjpro150 and copy your license key to ~/.mujoco/mjkey.txt
  2. pip install -r requirements.txt

Running

Configuration files can be found in config/.

Run

python main.py --config=config.ant

Acknowledgments

The code implementation is mainly modified based on the MBPO and AMPO codebase.