Fighting Fire with Fire: Using Antidote Data to Improve Polarization and Fairness of Recommender Systems

Abstract

The increasing role of recommender systems in many aspects of society makes it essential to consider how such systems may impact social good. Various modifications to recommendation algorithms have been proposed to improve their performance for specific socially relevant measures. However, previous proposals are often not easily adapted to different measures, and they generally require the ability to modify either existing system inputs, the system's algorithm, or the system's outputs. As an alternative, in this paper we introduce the idea of improving the social desirability of recommender system outputs by adding more data to the input, an approach we view as providing `antidote' data to the system. We formalize the antidote data problem, and develop optimization-based solutions. We take as our model system the matrix factorization approach to recommendation, and we propose a set of measures to capture the polarization or fairness of recommendations. We then show how to generate antidote data for each measure, pointing out a number of computational efficiencies, and discuss the impact on overall system accuracy. Our experiments show that a modest budget for antidote data can lead to significant improvements in the polarization or fairness of recommendations.

References to appendices are fixed

References in corpus (13)

• Equality of Opportunity in Supervised Learning
• Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatment
• Fairness of Exposure in Rankings
• FA*IR: A Fair Top-k Ranking Algorithm
• Biased Assimilation, Homophily and the Dynamics of Polarization
• How Algorithmic Confounding in Recommendation Systems Increases Homogeneity and Decreases Utility
• Fairness Constraints: Mechanisms for Fair Classification
• Equity of Attention: Amortizing Individual Fairness in Rankings
• Matrix Completion and Low-Rank SVD via Fast Alternating Least Squares
• Data Decisions and Theoretical Implications when Adversarially Learning Fair Representations
• A Convex Framework for Fair Regression
• Beyond Parity: Fairness Objectives for Collaborative Filtering
• Data Poisoning Attacks on Factorization-Based Collaborative Filtering

Cited by in corpus (16)

• Fairness in Machine Learning: A Survey
• A Survey on the Fairness of Recommender Systems
• Fairness in Recommender Systems: Research Landscape and Future Directions
• User Fairness, Item Fairness, and Diversity for Rankings in Two-Sided Markets
• Interplay between Upsampling and Regularization for Provider Fairness in Recommender Systems
• Equality of Learning Opportunity via Individual Fairness in Personalized Recommendations
• FairSR: Fairness-aware Sequential Recommendation through Multi-Task Learning with Preference Graph Embeddings
• Towards Individual and Multistakeholder Fairness in Tourism Recommender Systems
• RePBubLik: Reducing the Polarized Bubble Radius with Link Insertions
• Consumer Fairness in Recommender Systems: Contextualizing Definitions and Mitigations
• Leave No User Behind: Towards Improving the Utility of Recommender Systems for Non-mainstream Users
• Consumer-side Fairness in Recommender Systems: A Systematic Survey of Methods and Evaluation
• Intersectional Two-sided Fairness in Recommendation
• Adaptive Fair Representation Learning for Personalized Fairness in Recommendations via Information Alignment
• Fair Inputs and Fair Outputs: The Incompatibility of Fairness in Privacy and Accuracy
• Stairway to Fairness: Connecting Group and Individual Fairness