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

2026-01-24

AI Summary

This work addresses the suboptimality in cross-domain click-through rate (CTR) prediction caused by relying solely on domain-invariant features and target-domain-specific features. To overcome this limitation, the authors propose a novel mechanism that integrates non-aligned yet discriminative cross-domain features. Domain alignment is achieved through adversarial training optimized via Maximum Mean Discrepancy (MMD), while a dedicated feature disentanglement and reconstruction module separates domain-specific characteristics. The resulting representation fuses domain-invariant features, non-aligned discriminative features, and original contextual information to form a more comprehensive and enhanced embedding. Extensive experiments on real-world datasets and online A/B tests demonstrate that the proposed model significantly outperforms existing methods, confirming its effectiveness and practical utility in cross-domain CTR prediction.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Application Domains: Humanities & Computational Social Science
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Web Mining and Content Analysis: Large pretrained models with web data

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.

Problem

• Cross-domain recommendation
• CTR prediction
• Domain-invariant features
• Non-aligned features
• Data sparsity

Innovation

• adversarial alignment
• feature disentanglement
• cross-domain recommendation
• non-aligned features
• CTR prediction