LLM-based Federated Recommendation

2024-02-15

arXiv.org

AI Summary

To address the exacerbated client performance imbalance and excessive resource overhead in large language model (LLM)-based federated recommendation (Fed4Rec), this paper proposes PPLR, a privacy-preserving LLM federated recommendation framework. Methodologically, PPLR introduces two key innovations: (1) a novel dynamic parameter aggregation mechanism coupled with adaptive learning rate adjustment to mitigate training bias across heterogeneous clients; and (2) a hierarchical model partitioning strategy that retains sensitive layers locally while offloading non-sensitive layers to the server, substantially reducing on-device computational and storage demands. Experimental results demonstrate that PPLR preserves user behavioral data privacy while improving cross-client performance fairness and global recommendation accuracy. On average, it reduces client-side computation and storage overhead by over 30%.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Natural Language Processing: (Large) Language Models
• 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
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Large Language Models (LLMs), with their advanced contextual understanding abilities, have demonstrated considerable potential in enhancing recommendation systems via fine-tuning methods. However, fine-tuning requires users' behavior data, which poses considerable privacy risks due to the incorporation of sensitive user information. The unintended disclosure of such data could infringe upon data protection laws and give rise to ethical issues. To mitigate these privacy issues, Federated Learning for Recommendation (Fed4Rec) has emerged as a promising approach. Nevertheless, applying Fed4Rec to LLM-based recommendation presents two main challenges: first, an increase in the imbalance of performance across clients, affecting the system's efficiency over time, and second, a high demand on clients' computational and storage resources for local training and inference of LLMs. To address these challenges, we introduce a Privacy-Preserving LLM-based Recommendation (PPLR) framework. The PPLR framework employs two primary strategies. First, it implements a dynamic balance strategy, which involves the design of dynamic parameter aggregation and adjustment of learning speed for different clients during the training phase, to ensure relatively balanced performance across all clients. Second, PPLR adopts a flexible storage strategy, selectively retaining certain sensitive layers of the language model on the client side while offloading non-sensitive layers to the server. This approach aims to preserve user privacy while efficiently saving computational and storage resources. Experimental results demonstrate that PPLR not only achieves a balanced performance among clients but also enhances overall system performance in a manner that is both computationally and storage-efficient, while effectively protecting user privacy.

Problem

• Privacy risks in LLM-based recommendation systems
• Client performance imbalance in federated learning
• High client resource costs for LLM training

Innovation

• Federated Learning for LLM
• Dynamic balance strategy
• Flexible storage strategy