Towards Robust Cross-Domain Recommendation with Joint Identifiability of User Preference

TL;DR

This paper introduces a hierarchical modeling framework for cross-domain recommendation that ensures consistent user preference representations across domains through joint identifiability, improving robustness even with weak domain correlations.

Contribution

The paper proposes a novel hierarchical user preference modeling approach that enforces joint identifiability across domains, addressing limitations of disentanglement in cross-domain recommendation.

Findings

Outperforms state-of-the-art methods on real-world CDR tasks.

Effective even with weakly correlated domain pairs.

Highlights importance of joint identifiability for robustness.

Abstract

Recent cross-domain recommendation (CDR) studies assume that disentangled domain-shared and domain-specific user representations can mitigate domain gaps and facilitate effective knowledge transfer. However, achieving perfect disentanglement is challenging in practice, because user behaviors in CDR are highly complex, and the true underlying user preferences cannot be fully captured through observed user-item interactions alone. Given this impracticability, we instead propose to model {\it joint identifiability} that establishes unique correspondence of user representations across domains, ensuring consistent preference modeling even when user behaviors exhibit shifts in different domains. To achieve this, we introduce a hierarchical user preference modeling framework that organizes user representations by the neural network encoder's depth, allowing separate treatment of shallow and deeper subspaces. In the shallow subspace, our framework models the interest centroids for each user within each domain, probabilistically determining the users' interest belongings and selectively aligning these centroids across domains to ensure fine-grained consistency in domain-irrelevant features. For deeper subspace representations, we enforce joint identifiability by decomposing it into a shared cross-domain stable component and domain-variant components, linked by a bijective transformation for unique correspondence. Empirical studies on real-world CDR tasks with varying domain correlations demonstrate that our method consistently surpasses state-of-the-art, even with weakly correlated tasks, highlighting the importance of joint identifiability in achieving robust CDR.