Instrumental Variable Value Iteration for Causal Offline Reinforcement Learning


In offline reinforcement learning (RL) an optimal policy is learnt solely from a priori collected observational data. However, in observational data, actions are often confounded by unobserved variables. Instrumental variables (IVs), in the context of RL, are the variables whose influence on the state variables are all mediated through the action. When a valid instrument is present, we can recover the confounded transition dynamics through observational data. We study a confounded Markov decision process where the transition dynamics admit an additive nonlinear functional form. Using IVs, we derive a conditional moment restriction (CMR) through which we can identify transition dynamics based on observational data. We propose a provably efficient IV-aided Value Iteration (IVVI) algorithm based on a primal-dual reformulation of CMR. To the best of our knowledge, this is the first provably efficient algorithm for instrument-aided offline RL.

Technical report
Luofeng Liao
Luofeng Liao
MS Student (2020-2021)

Prior to graduate school, he received B.S. degree in Computer Science at Fudan University in June 2019. His research interests include high-dimensional statistis and distributed optimization. Luofeng continued his education as a PhD student at Columbia University. His personal website can be found here.

Mladen Kolar
Mladen Kolar
Associate Professor of Econometrics and Statistics

Mladen Kolar is an Associate Professor of Econometrics and Statistics at the University of Chicago Booth School of Business. His research is focused on high-dimensional statistical methods, graphical models, varying-coefficient models and data mining, driven by the need to uncover interesting and scientifically meaningful structures from observational data.