Automatic Feature Fairness in Recommendation via Adversaries

TL;DR

This paper introduces an adversarial training approach to enhance feature fairness in recommendation systems, balancing fairness and accuracy by dynamically adjusting perturbations based on feature biases.

Contribution

It proposes a novel adaptive adversarial feature learning method, AAFM, that improves fairness and generalization across diverse feature groups in recommendation models.

Findings

AAFM outperforms baselines in fairness and accuracy

Effective for both item- and user-fairness in various tasks

Adaptive perturbation improves model generalization

Abstract

Fairness is a widely discussed topic in recommender systems, but its practical implementation faces challenges in defining sensitive features while maintaining recommendation accuracy. We propose feature fairness as the foundation to achieve equitable treatment across diverse groups defined by various feature combinations. This improves overall accuracy through balanced feature generalizability. We introduce unbiased feature learning through adversarial training, using adversarial perturbation to enhance feature representation. The adversaries improve model generalization for under-represented features. We adapt adversaries automatically based on two forms of feature biases: frequency and combination variety of feature values. This allows us to dynamically adjust perturbation strengths and adversarial training weights. Stronger perturbations are applied to feature values with fewer combination varieties to improve generalization, while higher weights for low-frequency features address training imbalances. We leverage the Adaptive Adversarial perturbation based on the widely-applied Factorization Machine (AAFM) as our backbone model. In experiments, AAFM surpasses strong baselines in both fairness and accuracy measures. AAFM excels in providing item- and user-fairness for single- and multi-feature tasks, showcasing their versatility and scalability. To maintain good accuracy, we find that adversarial perturbation must be well-managed: during training, perturbations should not overly persist and their strengths should decay.