Adversarial Alignment and Disentanglement for Cross-Domain CTR Prediction with Domain-Encompassing Features

TL;DR

This paper introduces A^2DCDR, a novel cross-domain recommendation model that combines adversarial training and disentanglement to leverage both domain-invariant and non-aligned features, improving recommendation accuracy.

Contribution

The paper proposes a new model that captures comprehensive cross-domain information by integrating adversarial alignment, disentanglement, and fused representations, surpassing existing methods.

Findings

A^2DCDR outperforms baseline methods on real-world datasets.

The model demonstrates significant improvements in recommendation accuracy.

Online A/B testing confirms practical effectiveness.

Abstract

Cross-domain recommendation (CDR) has been increasingly explored to address data sparsity and cold-start issues. However, recent approaches typically disentangle domain-invariant features shared between source and target domains, as well as domain-specific features for each domain. However, they often rely solely on domain-invariant features combined with target domain-specific features, which can lead to suboptimal performance. To overcome the limitations, this paper presents the Adversarial Alignment and Disentanglement Cross-Domain Recommendation ($A^2DCDR$ ) model, an innovative approach designed to capture a comprehensive range of cross-domain information, including both domain-invariant and valuable non-aligned features. The $A^2DCDR$ model enhances cross-domain recommendation through three key components: refining MMD with adversarial training for better generalization, employing a feature disentangler and reconstruction mechanism for intra-domain disentanglement, and introducing a novel fused representation combining domain-invariant, non-aligned features with original contextual data. Experiments on real-world datasets and online A/B testing show that $A^2DCDR$ outperforms existing methods, confirming its effectiveness and practical applicability. The code is provided at https://github.com/youzi0925/A-2DCDR/tree/main.