Transfer Learning via Contextual Invariants for One-to-Many Cross-Domain Recommendation

Abstract

The rapid proliferation of new users and items on the social web has aggravated the gray-sheep user/long-tail item challenge in recommender systems. Historically, cross-domain co-clustering methods have successfully leveraged shared users and items across dense and sparse domains to improve inference quality. However, they rely on shared rating data and cannot scale to multiple sparse target domains (i.e., the one-to-many transfer setting). This, combined with the increasing adoption of neural recommender architectures, motivates us to develop scalable neural layer-transfer approaches for cross-domain learning. Our key intuition is to guide neural collaborative filtering with domain-invariant components shared across the dense and sparse domains, improving the user and item representations learned in the sparse domains. We leverage contextual invariances across domains to develop these shared modules, and demonstrate that with user-item interaction context, we can learn-to-learn informative representation spaces even with sparse interaction data. We show the effectiveness and scalability of our approach on two public datasets and a massive transaction dataset from Visa, a global payments technology company (19% Item Recall, 3x faster vs. training separate models for each domain). Our approach is applicable to both implicit and explicit feedback settings.

arXiv commentSIGIR 2020

References in corpus (11)

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Tutorial on Variational Autoencoders
• Unsupervised Domain Adaptation with Residual Transfer Networks
• Meta-SGD: Learning to Learn Quickly for Few-Shot Learning
• Translation-based Recommendation
• Online Batch Selection for Faster Training of Neural Networks
• Inf-VAE: A Variational Autoencoder Framework to Integrate Homophily and Influence in Diffusion Prediction
• Reducing Overfitting in Deep Networks by Decorrelating Representations
• Meta-Transfer Learning for Few-Shot Learning
• CnGAN: Generative Adversarial Networks for Cross-network user preference generation for non-overlapped users
• Sequential Scenario-Specific Meta Learner for Online Recommendation

Cited by in corpus (15)

• Robust Preference-Guided Denoising for Graph based Social Recommendation
• RecGURU: Adversarial Learning of Generalized User Representations for Cross-Domain Recommendation
• Towards Dynamic and Safe Configuration Tuning for Cloud Databases
• One for All, All for One: Learning and Transferring User Embeddings for Cross-Domain Recommendation
• Cross-Market Product Recommendation
• Review-Based Domain Disentanglement without Duplicate Users or Contexts for Cross-Domain Recommendation
• Doing More with Less: Overcoming Data Scarcity for POI Recommendation via Cross-Region Transfer
• User Invariant Preference Learning for Multi-Behavior Recommendation
• Better with Less: A Data-Active Perspective on Pre-Training Graph Neural Networks
• Semantic-enhanced Co-attention Prompt Learning for Non-overlapping Cross-Domain Recommendation
• Market-Aware Models for Efficient Cross-Market Recommendation
• Review-Based Hyperbolic Cross-Domain Recommendation
• Deep Unified Representation for Heterogeneous Recommendation
• CMML: Contextual Modulation Meta Learning for Cold-Start Recommendation
• Dual prototype attentive graph network for cross-market recommendation