Deep Jump Learning for Off-Policy Evaluation in Continuous Treatment Settings
This addresses the problem of evaluating treatment policies in continuous settings for fields like healthcare, representing an incremental advance by extending discrete methods to continuous domains.
The paper tackles off-policy evaluation for continuous treatments, such as personalized dosing, by developing a novel method that adaptively discretizes the treatment space using deep learning and change point detection, enabling application of existing discrete methods; results include theoretical justification, simulations, and a real application to Warfarin dosing.
We consider off-policy evaluation (OPE) in continuous treatment settings, such as personalized dose-finding. In OPE, one aims to estimate the mean outcome under a new treatment decision rule using historical data generated by a different decision rule. Most existing works on OPE focus on discrete treatment settings. To handle continuous treatments, we develop a novel estimation method for OPE using deep jump learning. The key ingredient of our method lies in adaptively discretizing the treatment space using deep discretization, by leveraging deep learning and multi-scale change point detection. This allows us to apply existing OPE methods in discrete treatments to handle continuous treatments. Our method is further justified by theoretical results, simulations, and a real application to Warfarin Dosing.