Graph-Sequential Alignment and Uniformity: Toward Enhanced Recommendation Systems

2024-12-05

arXiv.org

AI Summary

To address the signal fragmentation and suboptimal joint optimization between graph-structured and sequential recommendation paradigms, this paper proposes a cross-paradigm unified recommendation framework. The framework jointly trains a Graph Neural Network (GNN) for modeling high-order relational structures and a sequential model (e.g., Transformer) within a shared embedding space. We introduce a novel joint loss function—comprising cross-paradigm alignment and uniformity terms—that enables positive knowledge transfer between submodules. By eliminating feature mismatch common in multimodal fusion, our approach significantly enhances representation consistency and generalization capability. Extensive experiments on three real-world datasets demonstrate that the method consistently outperforms unimodal baselines and achieves state-of-the-art performance in recommendation. The source code is publicly available.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Game Theory and Economic Paradigms: Adversarial Learning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Graph-based and sequential methods are two popular recommendation paradigms, each excelling in its domain but lacking the ability to leverage signals from the other. To address this, we propose a novel method that integrates both approaches for enhanced performance. Our framework uses Graph Neural Network (GNN)-based and sequential recommenders as separate submodules while sharing a unified embedding space optimized jointly. To enable positive knowledge transfer, we design a loss function that enforces alignment and uniformity both within and across submodules. Experiments on three real-world datasets demonstrate that the proposed method significantly outperforms using either approach alone and achieves state-of-the-art results. Our implementations are publicly available at https://github.com/YuweiCao-UIC/GSAU.git.

Problem

• Recommendation Systems
• Graph Methods
• Sequence Methods

Innovation

• Graph Neural Networks
• Sequential Recommendation
• Shared Vocabulary Optimization