Privacy-Preserving Orthogonal Aggregation for Guaranteeing Gender Fairness in Federated Recommendation

TL;DR

This paper introduces PPOA, a privacy-preserving method for federated recommendation systems that enhances gender fairness and maintains user preferences while protecting sensitive attributes.

Contribution

It proposes a novel orthogonal aggregation technique combining secure schemes and quantization to improve fairness and privacy in federated recommendation systems.

Findings

PPOA improves recommendation accuracy for both genders by up to 8.25% and 6.36%.

PPOA achieves optimal fairness in most cases.

The method effectively maintains user preferences across gender groups.

Abstract

Under stringent privacy constraints, whether federated recommendation systems can achieve group fairness remains an inadequately explored question. Taking gender fairness as a representative issue, we identify three phenomena in federated recommendation systems: performance difference, data imbalance, and preference disparity. We discover that the state-of-the-art methods only focus on the first phenomenon. Consequently, their imposition of inappropriate fairness constraints detrimentally affects the model training. Moreover, due to insufficient sensitive attribute protection of existing works, we can infer the gender of all users with 99.90% accuracy even with the addition of maximal noise. In this work, we propose Privacy-Preserving Orthogonal Aggregation (PPOA), which employs the secure aggregation scheme and quantization technique, to prevent the suppression of minority groups by the majority and preserve the distinct preferences for better group fairness. PPOA can assist different groups in obtaining their respective model aggregation results through a designed orthogonal mapping while keeping their attributes private. Experimental results on three real-world datasets demonstrate that PPOA enhances recommendation effectiveness for both females and males by up to 8.25% and 6.36%, respectively, with a maximum overall improvement of 7.30%, and achieves optimal fairness in most cases. Extensive ablation experiments and visualizations indicate that PPOA successfully maintains preferences for different gender groups.