Skip to content

Implementation of paper "Residual Learning and Context Encoding for Adaptive Offline-to-Online RL"

Notifications You must be signed in to change notification settings

MohammadrezaNakhaei/ReLCE

Repository files navigation

ReLCE: Residual Learning and Context Encoding for Adaptive Offline-to-Online RL

Instructions

Install d4rl follwing this repo:

Install dependencies from requirements.txt

pip install -r rquirement.txt

Train the offline agent:

python train_offline.py --task hopper-medium-v2

Train residual agent:

python train.py --offline-path ./log/...

Citation

@article{ReLCE2024,
    title={ Residual Learning and Context Encoding for Adaptive Offline-to-Online Reinforcement Learning},
    author={Mohammadreza Nakhaei, Aidan Scannell, Joni Pajarinen}
    journal={},
    year={2024}
}

About

Implementation of paper "Residual Learning and Context Encoding for Adaptive Offline-to-Online RL"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages