CauseRec: Counterfactual User Sequence Synthesis for Sequential Recommendation

Abstract

Learning user representations based on historical behaviors lies at the core of modern recommender systems. Recent advances in sequential recommenders have convincingly demonstrated high capability in extracting effective user representations from the given behavior sequences. Despite significant progress, we argue that solely modeling the observational behaviors sequences may end up with a brittle and unstable system due to the noisy and sparse nature of user interactions logged. In this paper, we propose to learn accurate and robust user representations, which are required to be less sensitive to (attack on) noisy behaviors and trust more on the indispensable ones, by modeling counterfactual data distribution. Specifically, given an observed behavior sequence, the proposed CauseRec framework identifies dispensable and indispensable concepts at both the fine-grained item level and the abstract interest level. CauseRec conditionally samples user concept sequences from the counterfactual data distributions by replacing dispensable and indispensable concepts within the original concept sequence. With user representations obtained from the synthesized user sequences, CauseRec performs contrastive user representation learning by contrasting the counterfactual with the observational. We conduct extensive experiments on real-world public recommendation benchmarks and justify the effectiveness of CauseRec with multi-aspects model analysis. The results demonstrate that the proposed CauseRec outperforms state-of-the-art sequential recommenders by learning accurate and robust user representations.

Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2021)

References in corpus (13)

• Neural Graph Collaborative Filtering
• Session-based Recommendation with Graph Neural Networks
• GCC: Graph Contrastive Coding for Graph Neural Network Pre-Training
• S^3-Rec: Self-Supervised Learning for Sequential Recommendation with Mutual Information Maximization
• Personalizing Session-based Recommendations with Hierarchical Recurrent Neural Networks
• Temporal Relational Ranking for Stock Prediction
• Causal Embeddings for Recommendation
• Personalized News Recommendation with Context Trees
• Vista: A Visually, Socially, and Temporally-aware Model for Artistic Recommendation
• Estimating Position Bias without Intrusive Interventions
• General factorization framework for context-aware recommendations
• Should Graph Convolution Trust Neighbors? A Simple Causal Inference Method
• To Model or to Intervene: A Comparison of Counterfactual and Online Learning to Rank from User Interactions

Cited by in corpus (11)

• Filter-enhanced MLP is All You Need for Sequential Recommendation
• Debiased Contrastive Learning for Sequential Recommendation
• LinRec: Linear Attention Mechanism for Long-term Sequential Recommender Systems
• Should Graph Convolution Trust Neighbors? A Simple Causal Inference Method
• Graph Masked Autoencoder for Sequential Recommendation
• Contrastive Learning with Positive-Negative Frame Mask for Music Representation
• CT4Rec: Simple yet Effective Consistency Training for Sequential Recommendation
• Quantitatively Measuring and Contrastively Exploring Heterogeneity for Domain Generalization
• Calibration-Disentangled Learning and Relevance-Prioritized Reranking for Calibrated Sequential Recommendation
• PAS: A Position-Aware Similarity Measurement for Sequential Recommendation
• RecPFN: Prior-Fitted Networks for In-Context-Based Recommendations