Bilateral Unsymmetrical Graph Contrastive Learning for Recommendation

📄 arXiv: 2403.15075v1 📥 PDF

作者: Jiaheng Yu, Jing Li, Yue He, Kai Zhu, Shuyi Zhang, Wen Hu

分类: cs.IR, cs.AI

发布日期: 2024-03-22


💡 一句话要点

提出双边非对称图对比学习以解决推荐系统中的节点关系密度问题

🎯 匹配领域: 支柱二:RL算法与架构 (RL & Architecture)

关键词: 图对比学习 推荐系统 用户行为分析 深度学习 协同过滤

📋 核心要点

  1. 现有推荐方法未能有效处理用户和物品节点间关系密度的差异,限制了模型的推荐效果。
  2. 本文提出双边非对称图对比学习(BusGCL),通过双边切片对比训练优化用户和物品节点的关系建模。
  3. 在两个公共数据集上的实验表明,BusGCL在推荐性能上显著优于多种现有方法,展示了其有效性。

📝 摘要(中文)

近年来,研究者们利用图对比学习在图结构的用户-物品交互数据中进行协同过滤,已在推荐任务中展现出有效性。然而,现有方法忽视了用户和物品节点之间关系密度的差异,导致多跳图交互计算后图的适应性不同,从而限制了模型的性能。为此,本文提出了一种新颖的推荐框架——双边非对称图对比学习(BusGCL),通过双边切片对比训练,考虑用户-物品节点关系密度的非对称性,优化用户和物品图的推理能力。实验结果表明,BusGCL在多个推荐方法中表现出优越性,且其他模型可简单利用该方法提升推荐性能。

🔬 方法详解

问题定义:本文旨在解决现有推荐系统中用户和物品节点关系密度差异导致的适应性问题,现有方法未能充分利用图结构的潜力。

核心思路:提出双边非对称图对比学习(BusGCL),通过考虑用户和物品节点的非对称性,优化图推理过程,提升推荐效果。

技术框架:BusGCL框架包括三个主要模块:超图卷积网络(GCN)、标准GCN和扰动GCN,生成的嵌入分别被切片为用户和物品的子视图,并基于节点关系结构进行选择性组合。

关键创新:最重要的创新在于引入了双边切片对比训练,针对用户和物品节点的关系密度差异进行优化,这一设计与传统的对称图对比学习方法有本质区别。

关键设计:采用了分散损失函数来调整用户和物品嵌入的分布,确保嵌入之间的相互距离适当,从而保持学习能力。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

在两个公共数据集上的实验结果显示,BusGCL在推荐性能上显著优于多种基线方法,具体提升幅度达到XX%,证明了其有效性和实用性。

🎯 应用场景

该研究的潜在应用领域包括电子商务、社交网络和内容推荐等,能够有效提升推荐系统的性能,帮助用户更好地发现感兴趣的物品。未来,BusGCL方法可能在个性化推荐和用户行为分析等方面产生深远影响。

📄 摘要(原文)

Recent methods utilize graph contrastive Learning within graph-structured user-item interaction data for collaborative filtering and have demonstrated their efficacy in recommendation tasks. However, they ignore that the difference relation density of nodes between the user- and item-side causes the adaptability of graphs on bilateral nodes to be different after multi-hop graph interaction calculation, which limits existing models to achieve ideal results. To solve this issue, we propose a novel framework for recommendation tasks called Bilateral Unsymmetrical Graph Contrastive Learning (BusGCL) that consider the bilateral unsymmetry on user-item node relation density for sliced user and item graph reasoning better with bilateral slicing contrastive training. Especially, taking into account the aggregation ability of hypergraph-based graph convolutional network (GCN) in digging implicit similarities is more suitable for user nodes, embeddings generated from three different modules: hypergraph-based GCN, GCN and perturbed GCN, are sliced into two subviews by the user- and item-side respectively, and selectively combined into subview pairs bilaterally based on the characteristics of inter-node relation structure. Furthermore, to align the distribution of user and item embeddings after aggregation, a dispersing loss is leveraged to adjust the mutual distance between all embeddings for maintaining learning ability. Comprehensive experiments on two public datasets have proved the superiority of BusGCL in comparison to various recommendation methods. Other models can simply utilize our bilateral slicing contrastive learning to enhance recommending performance without incurring extra expenses.