Reinforcement Learning for Uplift Modeling

TL;DR

This paper introduces a novel reinforcement learning approach to uplift modeling by reformulating it as a Markov Decision Process, demonstrating significant improvements over existing methods through extensive experiments.

Contribution

The paper's main contribution is the reformulation of uplift modeling as an MDP, enabling the application of reinforcement learning techniques to improve treatment effect estimation.

Findings

Significant improvement over previous uplift modeling methods.

Effective application of reinforcement learning to individual treatment impact.

Validated on synthetic and real-world datasets.

Abstract

Uplift modeling aims to directly model the incremental impact of a treatment on an individual response. In this work, we address the problem from a new angle and reformulate it as a Markov Decision Process (MDP). We conducted extensive experiments on both a synthetic dataset and real-world scenarios, and showed that our method can achieve significant improvement over previous methods.