/SC-QD-MBPO

Primary LanguagePython

Model-based Policy Optimization with Trajectory Self-correction and Q-function Discriminator

This repository contains a TensorFlow implementation of [SC-DQ-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.