Invariant Policy Learning: A Causal Perspective

Sorawit Saengkyongam, Nikolaj Thams, Jonas Peters, Niklas Pfister

Research output: Contribution to journalJournal articleResearchpeer-review

3 Citations (Scopus)
14 Downloads (Pure)

Abstract

Contextual bandit and reinforcement learning algorithms have been successfully used in various interactive learning systems such as online advertising, recommender systems, and dynamic pricing. However, they have yet to be widely adopted in high-stakes application domains, such as healthcare. One reason may be that existing approaches assume that the underlying mechanisms are static in the sense that they do not change over different environments. In many real-world systems, however, the mechanisms are subject to shifts across environments which may invalidate the static environment assumption. In this paper, we take a step toward tackling the problem of environmental shifts considering the framework of offline contextual bandits. We view the environmental shift problem through the lens of causality and propose multi-environment contextual bandits that allow for changes in the underlying mechanisms. We adopt the concept of invariance from the causality literature and introduce the notion of policy invariance. We argue that policy invariance is only relevant if unobserved variables are present and show that, in that case, an optimal invariant policy is guaranteed to generalize across environments under suitable assumptions.

Original languageEnglish
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume45
Issue number7
Pages (from-to)8606-8620
Number of pages15
ISSN0162-8828
DOIs
Publication statusPublished - 2023

Bibliographical note

Publisher Copyright:
IEEE

Keywords

  • Causality
  • contextual bandits
  • distributional shift
  • Extraterrestrial measurements
  • Heuristic algorithms
  • off-policy learning
  • Particle measurements
  • Random variables
  • Reinforcement learning
  • Training
  • Visualization

Cite this