FedFlex: Federated Learning for Diverse Netflix Recommendations

2025-07-15

arXiv.org

AI Summary

In recommender systems, a fundamental tension exists among personalization, privacy preservation, and recommendation diversity. While federated learning (FL) enables model training without centralizing raw user data, its adverse impact on recommendation diversity remains unaddressed systematically. This paper proposes FedFlex, the first FL framework that jointly optimizes privacy, personalization, and diversity. It employs SVD- or BPR-based matrix factorization for personalized local model training and integrates Maximum Marginal Relevance (MMR) for client-side re-ranking to enhance diversity. Evaluated on a Netflix-style TV series recommendation scenario, FedFlex significantly improves content coverage and novelty of recommendation lists. A two-week online A/B test demonstrates sustained click-through rate (CTR), confirming that diversity enhancement does not compromise user engagement. The work thus establishes a principled approach to reconciling privacy, personalization, and diversity in federated recommendation.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Data Mining & Knowledge Management: Recommender Systems
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Decentralized Web and Fediverse systems

Abstract

Federated learning is a decentralized approach that enables collaborative model training across multiple devices while preserving data privacy. It has shown significant potential in various domains, including healthcare and personalized recommendation systems. However, most existing work on federated recommendation systems has focused primarily on improving accuracy, with limited attention to fairness and diversity. In this paper, we introduce FedFlex, a federated recommender system for Netflix-style TV series recommendations. FedFlex integrates two state-of-the-art matrix factorization algorithms for personalized fine-tuning. FedFlex also applies Maximal Marginal Relevance (MMR) to re-rank items and enhance diversity. We conduct extensive experiments comparing recommendations generated by SVD and BPR algorithms. In a live two-week user study, participants received two recommendation lists: List A, based on SVD or BPR, and List B, a re-ranked version emphasizing diversity. Participants were asked to click on the movies they were interested in watching. Our findings demonstrate that FedFlex effectively introduces diverse content, such as new genres, into recommendations without necessarily compromising user satisfaction.

Problem

• Federated learning addresses privacy-filter bubble tension in recommendations
• FedFlex combines local model fine-tuning with MMR re-ranking
• Framework evaluates diversity impact through live user study deployment

Innovation

• Local on-device fine-tuning of matrix factorization models
• Lightweight Maximal Marginal Relevance re-ranking step
• Two-stage framework combining personalization and diversity enhancement