Off-policy Evaluation in Infinite-Horizon Reinforcement Learning with Latent Confounders
This addresses a critical problem in high-stakes domains like healthcare and education where experimentation is limited and actions are confounded, though it is incremental by extending existing work to the confounded setting.
The paper tackles off-policy evaluation in infinite-horizon reinforcement learning with latent confounders by proposing a two-stage method that uses proxies to estimate stationary distribution ratios and combines them with optimal balancing, achieving theoretical consistency and empirical benchmarking.
Off-policy evaluation (OPE) in reinforcement learning is an important problem in settings where experimentation is limited, such as education and healthcare. But, in these very same settings, observed actions are often confounded by unobserved variables making OPE even more difficult. We study an OPE problem in an infinite-horizon, ergodic Markov decision process with unobserved confounders, where states and actions can act as proxies for the unobserved confounders. We show how, given only a latent variable model for states and actions, policy value can be identified from off-policy data. Our method involves two stages. In the first, we show how to use proxies to estimate stationary distribution ratios, extending recent work on breaking the curse of horizon to the confounded setting. In the second, we show optimal balancing can be combined with such learned ratios to obtain policy value while avoiding direct modeling of reward functions. We establish theoretical guarantees of consistency, and benchmark our method empirically.