Curious Replay for Model-based Adaptation
Abstract
Agents must be able to adapt quickly as an environment changes. We find that existing model-based reinforcement learning agents are unable to do this well, in part because of how they use past experiences to train their world model. Here, we present Curious Replay -- a form of prioritized experience replay tailored to model-based agents through use of a curiosity-based priority signal. Agents using Curious Replay exhibit improved performance in an exploration paradigm inspired by animal behavior and on the Crafter benchmark. DreamerV3 with Curious Replay surpasses state-of-the-art performance on Crafter, achieving a mean score of 19.4 that substantially improves on the previous high score of 14.5 by DreamerV3 with uniform replay, while also maintaining similar performance on the Deepmind Control Suite. Code for Curious Replay is available at https://github.com/AutonomousAgentsLab/curiousreplay
Cite
@article{arxiv.2306.15934,
title = {Curious Replay for Model-based Adaptation},
author = {Isaac Kauvar and Chris Doyle and Linqi Zhou and Nick Haber},
journal= {arXiv preprint arXiv:2306.15934},
year = {2023}
}
Comments
Accepted at ICML 2023. Website at https://sites.google.com/view/curious-replay