Learning Prescriptive ReLU Networks
This work addresses the need for interpretable and accurate policy learning in domains like healthcare or marketing, though it is incremental as it builds on existing neural network and tree methods.
The paper tackles the problem of learning optimal treatment policies from observational data by proposing a piecewise linear neural network model called P-ReLU, which achieves superior prescriptive accuracy compared to benchmarks and allows conversion into interpretable trees.
We study the problem of learning optimal policy from a set of discrete treatment options using observational data. We propose a piecewise linear neural network model that can balance strong prescriptive performance and interpretability, which we refer to as the prescriptive ReLU network, or P-ReLU. We show analytically that this model (i) partitions the input space into disjoint polyhedra, where all instances that belong to the same partition receive the same treatment, and (ii) can be converted into an equivalent prescriptive tree with hyperplane splits for interpretability. We demonstrate the flexibility of the P-ReLU network as constraints can be easily incorporated with minor modifications to the architecture. Through experiments, we validate the superior prescriptive accuracy of P-ReLU against competing benchmarks. Lastly, we present examples of interpretable prescriptive trees extracted from trained P-ReLUs using a real-world dataset, for both the unconstrained and constrained scenarios.