CUED Publications database

Deconfounding Reinforcement Learning in Observational Settings

Lu, C and Schölkopf, B and Hernández-Lobato, JM Deconfounding Reinforcement Learning in Observational Settings. (Unpublished)

Full text not available from this repository.

Abstract

We propose a general formulation for addressing reinforcement learning (RL) problems in settings with observational data. That is, we consider the problem of learning good policies solely from historical data in which unobserved factors (confounders) affect both observed actions and rewards. Our formulation allows us to extend a representative RL algorithm, the Actor-Critic method, to its deconfounding variant, with the methodology for this extension being easily applied to other RL algorithms. In addition to this, we develop a new benchmark for evaluating deconfounding RL algorithms by modifying the OpenAI Gym environments and the MNIST dataset. Using this benchmark, we demonstrate that the proposed algorithms are superior to traditional RL methods in confounded environments with observational data. To the best of our knowledge, this is the first time that confounders are taken into consideration for addressing full RL problems with observational data. Code is available at https://github.com/CausalRL/DRL.

Item Type: Article
Uncontrolled Keywords: cs.LG cs.LG stat.ML
Subjects: UNSPECIFIED
Divisions: Div F > Computational and Biological Learning
Depositing User: Cron Job
Date Deposited: 30 May 2019 01:15
Last Modified: 18 Aug 2020 12:52
DOI: