Abstract:Recommender systems have been widely deployed in many real-world applications, but usually suffer from the long-standing user cold-start problem. As a promising way, Cross-Domain Recommendation (CDR) has attracted a surge of inter-est, which aims to transfer the user preferences observed in the source domain to make recommendations in the target domain. Previous CDR approaches mostly achieve the goal by following the Embedding and Mapping (EMCDR) idea which attempts to learn a mapping function to transfer the pre-trained user repre-sentations (embeddings) from the source domain into the target domain. However, they pre-train the user/item representations independently for each domain, ignoring to consider both domain interactions simultaneously. Therefore, the biased pre-trained representations inevitably involve the domain-specific information which may lead to negative impact to transfer information across domains. In this work, we consider a key point of the CDR task: what information needs to be shared across domains? To achieve the above idea, this paper utilizes the information bottleneck (IB) principle, and proposes a novel approach termed as CDRIB to enforce the representations encoding the domain-shared information. To derive the unbiased representations, we devise two IB regularizers to model the cross-domain/in-domain user-item interactions simultaneously and thereby CDRIB could consider both domain interactions jointly for de-biasing. With an additional contrastive information regularizer, CDRIB can also capture cross-domain user-user correlations. In this way, those regularizers encourage the representations to encode the domain-shared information, which has the capability to make recommendations in both domains directly. To the best of our knowledge, this paper is the first work to capture the domain-shared information for cold-start users via variational information bottleneck. Empirical experiments illustrate that CDRIB outperforms the state-of-the-art approaches on four real-world cross-domain datasets, demonstrating the effectiveness of adopting the information bottleneck for CDR.

Not All Embeddings are Created Equal: Towards Robust Cross-domain Recommendation via Contrastive Learning

User Distribution Mapping Modelling with Collaborative Filtering for Cross Domain Recommendation

Contrastive Cross-domain Recommendation in Matching

A Unified Framework for Adaptive Representation Enhancement and Inversed Learning in Cross-Domain Recommendation

One for All, All for One: Learning and Transferring User Embeddings for Cross-Domain Recommendation

Beyond the Overlapping Users: Cross-Domain Recommendation via Adaptive Anchor Link Learning

Cross-Domain Recommendation to Cold-Start Users via Variational Information Bottleneck

Combining Graph Contrastive Embedding and Multi-head Cross-Attention Transfer for Cross-Domain Recommendation

DTCDR: A Framework for Dual-Target Cross-Domain Recommendation

Heterogeneous and Clustering-Enhanced Personalized Preference Transfer for Cross-Domain Recommendation

Decoupled domain-specific and domain-conditional representation learning for cross-domain recommendation

Mining User Consistent and Robust Preference for Unified Cross Domain Recommendation

Deep shared learning and attentive domain mapping for cross-domain recommendation

Contrastive Multi-View Interest Learning for Cross-Domain Sequential Recommendation

Exploiting Variational Domain-Invariant User Embedding for Partially Overlapped Cross Domain Recommendation

CausalCDR: Causal Embedding Learning for Cross-domain Recommendation

Gromov-Wasserstein Guided Representation Learning for Cross-Domain Recommendation.

Towards Universal Cross-Domain Recommendation

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

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

A Deep Dual Adversarial Network for Cross-domain Recommendation