Abstract:Sequential recommendation typically utilizes deep neural networks to mine rich information in interaction sequences. However, existing methods often face the issue of insufficient interaction data. To alleviate the sparsity issue, self-supervised learning is introduced into sequential recommendation. Despite its effectiveness, we argue that current self-supervised learning-based (i.e., SSL-based) sequential recommendation models have the following limitations: (1) using only a single self-supervised learning method, either contrastive self-supervised learning or generative self-supervised learning. (2) employing a simple data augmentation strategy in either the graph structure domain or the node feature domain. We believe that they have not fully utilized the capabilities of both self-supervised methods and have not sufficiently explored the advantages of combining graph augmentation schemes. As a result, they often fail to learn better item representations. In light of this, we propose a novel multi-task sequential recommendation framework named Adaptive Self-supervised Learning for sequential Recommendation (ASLRec). Specifically, our framework combines contrastive and generative self-supervised learning methods adaptively, simultaneously applying different perturbations at both the graph topology and node feature levels. This approach constructs diverse augmented graph views and employs multiple loss functions (including contrastive loss, generative loss, mask loss, and prediction loss) for joint training. By encompassing the capabilities of various methods, our model learns item representations across different augmented graph views to achieve better performance and effectively mitigate interaction noise and sparsity. In addition, we add a small proportion of random uniform noise to item representations, making the item representations more uniform and mitigating the inherent popularity bias in interaction records. We conduct extensive experiments on three publicly available benchmark datasets to evaluate our model. The results demonstrate that our approach achieves state-of-the-art performance compared to 14 other competitive methods: the hit rate (HR) improved by over 14.39%, and the normalized discounted cumulative gain (NDCG) increased by over 18.67%.

Graph Diffusion-based Representation Learning for Sequential Recommendation

Self-Propagation Graph Neural Network for Recommendation

Graph Attention Network with High-Order Neighbor Information Propagation for Social Recommendation

Dynamic Graph Neural Networks for Sequential Recommendation

Order-Aware Graph Neural Network for Sequential Recommendation

End-to-End Graph-Sequential Representation Learning for Accurate Recommendations

Edge-Enhanced Global Disentangled Graph Neural Network for Sequential Recommendation

Dual Sequential Recommendation Integrating High-order Collaborative Relations via Graph Attention Networks

Intention Enhanced Dual Heterogeneous Graph Attention Network for Sequential Recommendation

Learning Graph ODE for Continuous-Time Sequential Recommendation

Sequential Recommendation with Graph Neural Networks

A Novel Neighborhood-Augmented Graph Attention Network for Sequential Recommendation.

Instant Representation Learning for Recommendation over Large Dynamic Graphs

DiffuRec: A Diffusion Model for Sequential Recommendation

Denoising Item Graph with Disentangled Learning for Recommendation

Enhancing Sequential Recommendation with Graph Contrastive Learning

Multi-Grained Fusion Graph Neural Networks for Sequential Recommendation.

Graph Convolutional Networks with Collaborative Feature Fusion for Sequential Recommendation

Position-enhanced and Time-aware Graph Convolutional Network for Sequential Recommendations

Adaptive self-supervised learning for sequential recommendation

LDAG: Modeling Long-term Interests by Directed Acyclic Graph Neural Network for Sequential Recommendation