SYCAMORE-1 / ucb-MOPPO

implementation for UCB-driven Utility Function Search for Multi-objective Reinforcement Learning based on Decomposition

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

To run the proposed method and baselines on Mujoco V2, you need to run on linux kernal, 
follow the command.txt to build a independed docker environment on windows.

To run the proposed method on lasted Mujoco-V4, pip install gymnasium and follow the lastest apis on 
https://gymnasium.farama.org/environments/mujoco/

to run proposed methods on different problems with different seeds:
python ucb.py --eidx 0 --sidx 0 

About

implementation for UCB-driven Utility Function Search for Multi-objective Reinforcement Learning based on Decomposition


Languages

Language:Python 99.1%Language:Dockerfile 0.9%