Delphic Offline Reinforcement Learning under Nonidentifiable Hidden Confounding
Open access
Date
2023-06-01Type
- Working Paper
ETH Bibliography
yes
Altmetrics
Abstract
A prominent challenge of offline reinforcement learning (RL) is the issue of hidden confounding: unobserved variables may influence both the actions taken by the agent and the observed outcomes. Hidden confounding can compromise the validity of any causal conclusion drawn from data and presents a major obstacle to effective offline RL. In the present paper, we tackle the problem of hidden confounding in the nonidentifiable setting. We propose a definition of uncertainty due to hidden confounding bias, termed delphic uncertainty, which uses variation over world models compatible with the observations, and differentiate it from the well-known epistemic and aleatoric uncertainties. We derive a practical method for estimating the three types of uncertainties, and construct a pessimistic offline RL algorithm to account for them. Our method does not assume identifiability of the unobserved confounders, and attempts to reduce the amount of confounding bias. We demonstrate through extensive experiments and ablations the efficacy of our approach on a sepsis management benchmark, as well as on electronic health records. Our results suggest that nonidentifiable hidden confounding bias can be mitigated to improve offline RL solutions in practice. Show more
Permanent link
https://doi.org/10.3929/ethz-b-000638581Publication status
publishedJournal / series
arXivPages / Article No.
Publisher
Cornell UniversityEdition / version
v1Organisational unit
09568 - Rätsch, Gunnar / Rätsch, Gunnar
02219 - ETH AI Center / ETH AI Center
More
Show all metadata
ETH Bibliography
yes
Altmetrics