Contrastive Meta Learning with Behavior Multiplicity for Recommendation

Abstract

A well-informed recommendation framework could not only help users identify their interested items, but also benefit the revenue of various online platforms (e.g., e-commerce, social media). Traditional recommendation models usually assume that only a single type of interaction exists between user and item, and fail to model the multiplex user-item relationships from multi-typed user behavior data, such as page view, add-to-favourite and purchase. While some recent studies propose to capture the dependencies across different types of behaviors, two important challenges have been less explored: i) Dealing with the sparse supervision signal under target behaviors (e.g., purchase). ii) Capturing the personalized multi-behavior patterns with customized dependency modeling. To tackle the above challenges, we devise a new model CML, Contrastive Meta Learning (CML), to maintain dedicated crosstype behavior dependency for different users. In particular, we propose a multi-behavior contrastive learning framework to distill transferable knowledge across different types of behaviors via the constructed contrastive loss. In addition, to capture the diverse multi-behavior patterns, we design a contrastive meta network to encode the customized behavior heterogeneity for different users. Extensive experiments on three real-world datasets indicate that our method consistently outperforms various state-of-the-art recommendation methods. Our empirical studies further suggest that the contrastive meta learning paradigm offers great potential for capturing the behavior multiplicity in recommendation. We release our model implementation at: https://github.com/weiwei1206/CML.git. CCS CONCEPTS• Information systems → Recommender systems.

Knowledge Graph Contrastive Learning for Recommendation

“…It aims to learn quality discriminative representations by contrasting positive and negative samples from different views. Several recent attempts have brought the self-supervised learning to the recommendation [21,22,43,44]. For example, SGL [44] performs dropout operations over the graph connection structures with different strategies, i.e.,, node dropout, edge dropout and random walk.…”

“…For example, SGL [44] performs dropout operations over the graph connection structures with different strategies, i.e.,, node dropout, edge dropout and random walk. Additionally, CML [43] enhances the recommender system with the consideration of multi-behavior relationships between users and items with contrastive learning. Motivated by these existing contrastive learning frameworks, this work develops a new graph contrastive learning paradigm for recommendation by effectively integrating knowledge graph representation and user-item interaction augmentation.…”

Hypergraph Contrastive Collaborative Filtering

“…Contrastive learning has become an effective self-supervised framework, to capture the feature representation consistency under different views [27,37]. It has achieved promising performance in various domains, such as visual data representation [5,28], language data understanding [2,31], graph representation learning [29,51] and recommender systems [22,38,45,46].…”

Spatial-Temporal Hypergraph Self-Supervised Learning for Crime Prediction

“…In self-supervised learning paradigms, models explore the supervision signals from the data itself with auxiliary learning tasks. Furthermore, contrastive-based SSL methods aim to reach agreement between generated correlated contrastive views [39]. However, self-supervised learning is relatively less explored in spatialtemporal data prediction.…”