Scalable and Provably Fair Exposure Control for Large-Scale Recommender Systems
Riku Togashi, Kenshi Abe, Yuta Saito
TL;DR
Exposure fairness in recommender systems is essential but challenging at industry scale. The authors introduce exADMM, an ADMM-based extension of iALS that incorporates a fairness regularizer via an auxiliary variable to decouple dependencies and enable scalable optimization. They provide convergence guarantees and demonstrate through experiments that exADMM achieves flexible accuracy–fairness tradeoffs with competitive performance and superior scalability compared to existing fair-recommendation methods. The work has practical impact by making provably fair exposure control feasible for systems with millions of users and items, addressing real-world deployment needs and societal concerns.
Abstract
Typical recommendation and ranking methods aim to optimize the satisfaction of users, but they are often oblivious to their impact on the items (e.g., products, jobs, news, video) and their providers. However, there has been a growing understanding that the latter is crucial to consider for a wide range of applications, since it determines the utility of those being recommended. Prior approaches to fairness-aware recommendation optimize a regularized objective to balance user satisfaction and item fairness based on some notion such as exposure fairness. These existing methods have been shown to be effective in controlling fairness, however, most of them are computationally inefficient, limiting their applications to only unrealistically small-scale situations. This indeed implies that the literature does not yet provide a solution to enable a flexible control of exposure in the industry-scale recommender systems where millions of users and items exist. To enable a computationally efficient exposure control even for such large-scale systems, this work develops a scalable, fast, and fair method called \emph{\textbf{ex}posure-aware \textbf{ADMM} (\textbf{exADMM})}. exADMM is based on implicit alternating least squares (iALS), a conventional scalable algorithm for collaborative filtering, but optimizes a regularized objective to achieve a flexible control of accuracy-fairness tradeoff. A particular technical challenge in developing exADMM is the fact that the fairness regularizer destroys the separability of optimization subproblems for users and items, which is an essential property to ensure the scalability of iALS. Therefore, we develop a set of optimization tools to enable yet scalable fairness control with provable convergence guarantees as a basis of our algorithm.