Stateful Offline Contextual Policy Evaluation and Learning
This addresses offline policy optimization for operations management problems like pricing, but it is incremental as it builds on existing contextual bandit and MDP frameworks.
The paper tackles off-policy evaluation and learning in Markov decision processes with exogenous arrivals, such as dynamic personalized pricing, by leveraging causal structure to generalize individual-level responses across timesteps. It shows improved out-of-sample policy performance in simulations, with analysis of sample complexity and error amplification.
We study off-policy evaluation and learning from sequential data in a structured class of Markov decision processes that arise from repeated interactions with an exogenous sequence of arrivals with contexts, which generate unknown individual-level responses to agent actions. This model can be thought of as an offline generalization of contextual bandits with resource constraints. We formalize the relevant causal structure of problems such as dynamic personalized pricing and other operations management problems in the presence of potentially high-dimensional user types. The key insight is that an individual-level response is often not causally affected by the state variable and can therefore easily be generalized across timesteps and states. When this is true, we study implications for (doubly robust) off-policy evaluation and learning by instead leveraging single time-step evaluation, estimating the expectation over a single arrival via data from a population, for fitted-value iteration in a marginal MDP. We study sample complexity and analyze error amplification that leads to the persistence, rather than attenuation, of confounding error over time. In simulations of dynamic and capacitated pricing, we show improved out-of-sample policy performance in this class of relevant problems.