A General Theory of Outcome Weighted Learning for Individualized Treatment Rules

TL;DR

This paper develops a comprehensive theoretical framework for Outcome Weighted Learning (OWL) using Matern kernels, providing convergence guarantees and new algorithms for personalized treatment rule estimation.

Contribution

It introduces a general relationship between 0-1 risk and surrogate losses, extends OWL theory to Matern kernels, and proposes new reweighted optimization algorithms.

Findings

Strong performance in simulations and ACTG 175 application.

Convergence rates established for kernel-based OWL.

Framework accommodates nonconvex losses and Matern kernels.

Abstract

Personalized medicine aims to tailor treatments to individual patients, especially when people respond heterogeneously to therapies. A key objective is to learn individualized treatment rules that recommend optimal treatments from patient characteristics. Outcome weighted learning (OWL) is an important framework because it reformulates the task as a weighted classification problem targeting clinical benefit and using modern machine learning tools. Existing OWL theory has been focusing on specific surrogate losses and Gaussian kernels. Matern kernels, which allow adjustable smoothness and better match many real world data structures, are often more suitable and include the Gaussian kernel as a special case. This work develops a general relationship between population 0-1 risk and risks from a broad class of nonnegative surrogate losses using a constrained variational transformation. The transform simplifies for convex losses and provides simple expressions for certain nonconvex losses. A condition is established that ensures a nontrivial upper bound on the excess 0-1 risk. The paper establishes convergence rates for kernel based OWL under smoothness conditions with Matern kernels or geometric noise conditions with Gaussian kernels for both convex and nonconvex losses. It also proposes two iteratively reweighted convex optimization algorithms. Simulations and an application to ACTG 175 show strong performance.