Code for the MRS.Q algorithm from The Surprising Difficulty of Search in Model-Based Reinforcement Learning by Wei-Di Chang, Mikael Henaff, Brandon Amos, Gregory Dudek, and Scott Fujimoto.
Benchmark is designated by a prefix (Gym-, Dmc-, HumanoidBench-) followed by the original environment name. A complete list of environments are contained in MRSQ/utils.py.
Example usage:
cd MRSQ
python main.py --env Gym-HalfCheetah-v4
python main.py --env Dmc-quadruped-walk
- Agent, networks, hyperparameters: MRSQ/MRSQ.py.
- Replay buffer: MRSQ/buffer.py.
- Environment preprocesing: MRSQ/env_preprocessing.py.
Results are formatted in human-readable .txt files under /results. There is a code snippet in MRSQ/utils.py to process the .txt files into a dictionary of arrays.
MRSQ is licensed under the CC BY-NC 4.0 license, as found in the LICENSE file.