Efficient Real-world Testing of Causal Decision Making via Bayesian Experimental Design for Contextual Optimisation
This addresses the need for efficient testing of personalised treatments in applications like customer targeting, though it appears incremental as it builds on existing Bayesian methods.
The paper tackles the problem of real-world testing for causal decision making by introducing a model-agnostic framework using Bayesian Experimental Design to evaluate and improve contextual treatment assignments, achieving superior performance in simulation studies compared to baselines.
The real-world testing of decisions made using causal machine learning models is an essential prerequisite for their successful application. We focus on evaluating and improving contextual treatment assignment decisions: these are personalised treatments applied to e.g. customers, each with their own contextual information, with the aim of maximising a reward. In this paper we introduce a model-agnostic framework for gathering data to evaluate and improve contextual decision making through Bayesian Experimental Design. Specifically, our method is used for the data-efficient evaluation of the regret of past treatment assignments. Unlike approaches such as A/B testing, our method avoids assigning treatments that are known to be highly sub-optimal, whilst engaging in some exploration to gather pertinent information. We achieve this by introducing an information-based design objective, which we optimise end-to-end. Our method applies to discrete and continuous treatments. Comparing our information-theoretic approach to baselines in several simulation studies demonstrates the superior performance of our proposed approach.