Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Unbiased and Robust: External Attention-enhanced Graph Contrastive Learning for Cross-domain Sequential Recommendation

Xinhua Wang, Houping Yue, Zizheng Wang, Liancheng Xu, Jinyu Zhang

TL;DR

An External Attention-enhanced Graph Contrastive Learning framework, namely EA-GCL, is proposed that outperforms several state-of-the-art baselines on CSR tasks and can effectively alleviate the bias interference from the batch-based training scheme.

Abstract

Cross-domain sequential recommenders (CSRs) are gaining considerable research attention as they can capture user sequential preference by leveraging side information from multiple domains. However, these works typically follow an ideal setup, i.e., different domains obey similar data distribution, which ignores the bias brought by asymmetric interaction densities (a.k.a. the inter-domain density bias). Besides, the frequently adopted mechanism (e.g., the self-attention network) in sequence encoder only focuses on the interactions within a local view, which overlooks the global correlations between different training batches. To this end, we propose an External Attention-enhanced Graph Contrastive Learning framework, namely EA-GCL. Specifically, to remove the impact of the inter-domain density bias, an auxiliary Self-Supervised Learning (SSL) task is attached to the traditional graph encoder under a multi-task learning manner. To robustly capture users' behavioral patterns, we develop an external attention-based sequence encoder that contains an MLP-based memory-sharing structure. Unlike the self-attention mechanism, such a structure can effectively alleviate the bias interference from the batch-based training scheme. Extensive experiments on two real-world datasets demonstrate that EA-GCL outperforms several state-of-the-art baselines on CSR tasks. The source codes and relevant datasets are available at https://github.com/HoupingY/EA-GCL.

Unbiased and Robust: External Attention-enhanced Graph Contrastive Learning for Cross-domain Sequential Recommendation

TL;DR

An External Attention-enhanced Graph Contrastive Learning framework, namely EA-GCL, is proposed that outperforms several state-of-the-art baselines on CSR tasks and can effectively alleviate the bias interference from the batch-based training scheme.

Abstract

Cross-domain sequential recommenders (CSRs) are gaining considerable research attention as they can capture user sequential preference by leveraging side information from multiple domains. However, these works typically follow an ideal setup, i.e., different domains obey similar data distribution, which ignores the bias brought by asymmetric interaction densities (a.k.a. the inter-domain density bias). Besides, the frequently adopted mechanism (e.g., the self-attention network) in sequence encoder only focuses on the interactions within a local view, which overlooks the global correlations between different training batches. To this end, we propose an External Attention-enhanced Graph Contrastive Learning framework, namely EA-GCL. Specifically, to remove the impact of the inter-domain density bias, an auxiliary Self-Supervised Learning (SSL) task is attached to the traditional graph encoder under a multi-task learning manner. To robustly capture users' behavioral patterns, we develop an external attention-based sequence encoder that contains an MLP-based memory-sharing structure. Unlike the self-attention mechanism, such a structure can effectively alleviate the bias interference from the batch-based training scheme. Extensive experiments on two real-world datasets demonstrate that EA-GCL outperforms several state-of-the-art baselines on CSR tasks. The source codes and relevant datasets are available at https://github.com/HoupingY/EA-GCL.
Paper Structure (24 sections, 16 equations, 4 figures, 2 tables)

This paper contains 24 sections, 16 equations, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Sequential Recommendation
  4. Cross-domain Sequential Recommendation
  5. Unbiased Recommendation
  6. Method
  7. Preliminary
  8. Overview of EA-GCL
  9. Node Representation Learning
  10. Graph Contrastive Learning
  11. Data Augmentation on Graph Structure
  12. Contrastive Learning
  13. Sequence Representation Learning
  14. Prediction Layers & Loss Functions
  15. Experiment
  16. ...and 9 more sections

Figures (4)

  • Figure 1: shows the overview of EA-GCL, where $\bm{S}_1$ to $\bm{S}_4$ denote the hybrid interaction sequences of two overlapped users $\bm{U}_1$ and $\bm{U}_2$.
  • Figure 2: shows the details of the external attention-based sequence encoder (take domain $\mathcal{A}$ as an example here).
  • Figure 3: Impact of hyper-parameter $\alpha$ and $\beta$ (take DOUBAN dataset as an example).
  • Figure 4: Model Training Efficiency on DOUBAN and AMAZON.