Factored Adaptation for Non-Stationary Reinforcement Learning

Open Access
Authors
Publication date 2023
Host editors
  • S. Koyejo
  • S. Mohamed
  • A. Agarwal
  • D. Belgrave
  • K. Cho
  • A. Oh
Book title 36th Conference on Neural Information Processing Systems (NeurIPS 2022)
Book subtitle New Orleans, Louisiana, USA, 28 November-9 December 2022
ISBN
  • 9781713871088
ISBN (electronic)
  • 9781713873129
Series Advances in Neural Information Processing Systems
Event Thirty-sixth Conference on Neural Information Processing Systems
Volume | Issue number 41
Pages (from-to) 31957-31971
Publisher San Diego, CA: Neural Information Processing Systems Foundation
Organisations
  • Faculty of Science (FNWI) - Informatics Institute (IVI)
Abstract
Dealing with non-stationarity in environments (e.g., in the transition dynamics) and objectives (e.g., in the reward functions) is a challenging problem that is crucial in real-world applications of reinforcement learning (RL). While most current approaches model the changes as a single shared embedding vector, we leverage insights from the recent causality literature to model non-stationarity in terms of individual latent change factors, and causal graphs across different environments. In particular, we propose Factored Adaptation for Non-Stationary RL (FANS-RL), a factored adaption approach that learns jointly both the causal structure in terms of a factored MDP, and a factored representation of the individual time-varying change factors. We prove that under standard assumptions, we can completely recover the causal graph representing the factored transition and reward function, as well as a partial structure between the individual change factors and the state components. Through our general framework, we can consider general non-stationary scenarios with different function types and changing frequency, including changes across episodes and within episodes. Experimental results demonstrate that FANS-RL outperforms existing approaches in terms of return, compactness of the latent state representation, and robustness to varying degrees of non-stationarity.
Document type Conference contribution
Note With supplemental file
Language English
Published at https://doi.org/10.48550/arXiv.2203.16582
Published at https://papers.nips.cc/paper_files/paper/2022/hash/cf4356f994917177213c55ff438ddf71-Abstract-Conference.html
Other links https://www.proceedings.com/68431.html
Downloads
2203.16582_with appendix (Accepted author manuscript)
Supplementary materials
Permalink to this page
Back