Adaptive Coordinators and Prompts on Heterogeneous Graphs for Cross-Domain Recommendations

2024-10-15

arXiv.org

AI Summary

Cross-domain recommendation faces dual challenges: difficulty in fusing multi-source heterogeneous graph data and high risk of negative transfer. To address these, we propose HAGO—a Hierarchical Adaptive Graph Orchestration framework. HAGO introduces a heterogeneous adaptive graph coordinator that dynamically integrates interaction graphs from multiple domains, and incorporates a learnable graph prompting mechanism to enable controllable knowledge transfer during unified multi-domain graph pretraining. Its core innovations include: (i) the first heterogeneous graph structure-adaptive orchestration architecture; (ii) a multi-domain joint pretraining strategy; and (iii) a graph-prompt-driven embedding adaptation method—collectively mitigating negative transfer. Extensive experiments demonstrate that HAGO consistently outperforms state-of-the-art methods across multiple cross-domain benchmarks, while maintaining strong generalizability and deployment feasibility.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Application Domains: Humanities & Computational Social Science
• Multiagent Systems: Adversarial Agents

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

In the online digital world, users frequently engage with diverse items across multiple domains (e.g., e-commerce platforms, streaming services, and social media networks), forming complex heterogeneous interaction graphs. Leveraging this multi-domain information can undoubtedly enhance the performance of recommendation systems by providing more comprehensive user insights and alleviating data sparsity in individual domains. However, integrating multi-domain knowledge for the cross-domain recommendation is very hard due to inherent disparities in user behavior and item characteristics and the risk of negative transfer, where irrelevant or conflicting information from the source domains adversely impacts the target domain's performance. To address these challenges, we offer HAGO, a novel framework with $ extbf{H}$eterogeneous $ extbf{A}$daptive $ extbf{G}$raph co$ extbf{O}$rdinators, which dynamically integrate multi-domain graphs into a cohesive structure by adaptively adjusting the connections between coordinators and multi-domain graph nodes, thereby enhancing beneficial inter-domain interactions while mitigating negative transfer effects. Additionally, we develop a universal multi-domain graph pre-training strategy alongside HAGO to collaboratively learn high-quality node representations across domains. To effectively transfer the learned multi-domain knowledge to the target domain, we design an effective graph prompting method, which incorporates pre-trained embeddings with learnable prompts for the recommendation task. Our framework is compatible with various graph-based models and pre-training techniques, demonstrating broad applicability and effectiveness. Further experimental results show that our solutions outperform state-of-the-art methods in multi-domain recommendation scenarios and highlight their potential for real-world applications.

Problem

• Integrating multi-domain data for cross-domain recommendation
• Addressing disparities in user behavior and item characteristics
• Mitigating negative transfer from irrelevant source domains

Innovation

• Heterogeneous Adaptive Graph Coordinators for integration
• Dynamic multi-domain graph cohesive structure
• Universal multi-domain graph pre-training strategy