Multifaceted User Modeling in Recommendation: A Federated Foundation Models Approach

TL;DR

This paper introduces a novel federated Transformer-based recommendation model that captures multifaceted user interests through a specialized layer and group gating, improving performance while preserving privacy.

Contribution

It presents a new Transformer layer with group gating for hierarchical multifaceted user modeling and extends the framework to federated learning for privacy-preserving recommendations.

Findings

Superior performance on benchmark datasets

Effective capture of diverse user interests

Enhanced privacy through federated learning

Abstract

Multifaceted user modeling aims to uncover fine-grained patterns and learn representations from user data, revealing their diverse interests and characteristics, such as profile, preference, and personality. Recent studies on foundation model-based recommendation have emphasized the Transformer architecture's remarkable ability to capture complex, non-linear user-item interaction relationships. This paper aims to advance foundation model-based recommendersystems by introducing enhancements to multifaceted user modeling capabilities. We propose a novel Transformer layer designed specifically for recommendation, using the self-attention mechanism to capture sequential user-item interaction patterns. Specifically, we design a group gating network to identify user groups, enabling hierarchical discovery across different layers, thereby capturing the multifaceted nature of user interests through multiple Transformer layers. Furthermore, to broaden the data scope and further enhance multifaceted user modeling, we extend the framework to a federated setting, enabling the use of private datasets while ensuring privacy. Experimental validations on benchmark datasets demonstrate the superior performance of our proposed method. Code is available.