Robustness-enhanced Uplift Modeling with Adversarial Feature Desensitization

TL;DR

This paper introduces RUAD, a novel framework that enhances the robustness of uplift models in online marketing by reducing their sensitivity to key feature perturbations through adversarial training and feature selection.

Contribution

The paper proposes a new robustness-enhanced uplift modeling framework with adversarial feature desensitization, addressing feature sensitivity issues in practical applications.

Findings

RUAD effectively reduces feature sensitivity in uplift models.

Experimental results show improved robustness on real-world datasets.

RUAD is compatible with various uplift models.

Abstract

Uplift modeling has shown very promising results in online marketing. However, most existing works are prone to the robustness challenge in some practical applications. In this paper, we first present a possible explanation for the above phenomenon. We verify that there is a feature sensitivity problem in online marketing using different real-world datasets, where the perturbation of some key features will seriously affect the performance of the uplift model and even cause the opposite trend. To solve the above problem, we propose a novel robustness-enhanced uplift modeling framework with adversarial feature desensitization (RUAD). Specifically, our RUAD can more effectively alleviate the feature sensitivity of the uplift model through two customized modules, including a feature selection module with joint multi-label modeling to identify a key subset from the input features and an adversarial feature desensitization module using adversarial training and soft interpolation operations to enhance the robustness of the model against this selected subset of features. Finally, we conduct extensive experiments on a public dataset and a real product dataset to verify the effectiveness of our RUAD in online marketing. In addition, we also demonstrate the robustness of our RUAD to the feature sensitivity, as well as the compatibility with different uplift models.