Cluster-driven Personalized Federated Recommendation with Interest-aware Graph Convolution Network for Multimedia
Abstract: Federated learning addresses privacy concerns in multimedia recommender systems by enabling collaborative model training without exchanging raw data. However, existing federated recommendation models are mainly based on basic backbones like Matrix Factorization (MF), which are inadequate to capture complex implicit interactions between users and multimedia content. Graph Convolutional Networks (GCNs) offer a promising method by utilizing the information from high-order neighbors, but face challenges in federated settings due to problems such as over-smoothing, data heterogeneity, and elevated communication expenses. To resolve these problems, we propose a Cluster-driven Personalized Federated Recommender System with Interest-aware Graph Convolution Network (CPF-GCN) for multimedia recommendation. CPF-GCN comprises a local interest-aware GCN module that optimizes node representations through subgraph-enhanced adaptive graph convolution operations, mitigating the over-smoothing problem by adaptively extracting information from layers and selectively utilizing high-order connectivity based on user interests. Simultaneously, a cluster-driven aggregation approach at the server significantly reduces communication costs by selectively aggregating models from clusters. The aggregation produces a global model and cluster-level models, combining them with the user's local model allows us to tailor the recommendation model for the user, achieving personalized recommendations. Moreover, we propose an adversarial optimization technique to further augment the robustness of CPF-GCN. Experiments on three datasets demonstrate that CPF-GCN significantly outperforms the state-of-the-art models.
Primary Subject Area: [Engagement] Multimedia Search and Recommendation
Relevance To Conference: As multimedia content grows exponentially, finding relevant and appealing content becomes increasingly challenging for users. In multimedia content sharing platforms like Instagram, YouTube, and TikTok, multimedia recommendation has become a core service to help users identify content of interest. These recommender systems have become a key hub, connecting users, content, and business. They play an important role in improving user experience, facilitating content dissemination, and innovating business models. Even though traditional recommender systems are helpful, they often need to collect extensive personal data, posing significant privacy risks. Federated learning approaches mitigate privacy concerns, but they face the issue of data heterogeneity and struggle to achieve personalized recommendations. To address these challenges, we propose a Cluster-driven Personalized Federated Recommender System with Interest-aware Graph Convolution Network (CPF-GCN), which can effectively safeguard user privacy while delivering accurate personalized multimedia recommendations. Overall, our work contributes to the multimedia field by addressing key challenges in privacy-preserving and personalized multimedia recommendations, offering a promising solution that balances user privacy with the need for accurate multimedia content discovery. We are confident that our work can significantly enhance the user experience and business value of multimedia platforms, while respecting individual privacy rights.
Submission Number: 1919
Loading