Deconfounded Causal Collaborative Filtering

Abstract

Recommender systems may be confounded by various types of confounding factors (also called confounders) that may lead to inaccurate recommendations and sacrificed recommendation performance. Current approaches to solving the problem usually design each specific model for each specific confounder. However, real-world systems may include a huge number of confounders and thus designing each specific model for each specific confounder could be unrealistic. More importantly, except for those ``explicit confounders'' that experts can manually identify and process such as item's position in the ranking list, there are also many ``latent confounders'' that are beyond the imagination of experts. For example, users' rating on a song may depend on their current mood or the current weather, and users' preference on ice creams may depend on the air temperature. Such latent confounders may be unobservable in the recorded training data. To solve the problem, we propose Deconfounded Causal Collaborative Filtering (DCCF). We first frame user behaviors with unobserved confounders into a causal graph, and then we design a front-door adjustment model carefully fused with machine learning to deconfound the influence of unobserved confounders. Experiments on real-world datasets show that our method is able to deconfound unobserved confounders to achieve better recommendation performance.

Accepted by the ACM Transactions on Recommender Systems (TORS)

References in corpus (16)

• BPR: Bayesian Personalized Ranking from Implicit Feedback
• Towards Long-term Fairness in Recommendation
• Collaborative Filtering and the Missing at Random Assumption
• Deconfounded Recommendation for Alleviating Bias Amplification
• Counterfactual Explainable Recommendation
• Estimating Position Bias without Intrusive Interventions
• Learning and Evaluating Graph Neural Network Explanations based on Counterfactual and Factual Reasoning
• Cascade Model-based Propensity Estimation for Counterfactual Learning to Rank
• Multi-sided Exposure Bias in Recommendation
• Counterfactual Collaborative Reasoning
• Dynamic Causal Collaborative Filtering
• Causal Inference for Recommendation: Foundations, Methods and Applications
• Mitigating Hidden Confounding Effects for Causal Recommendation
• Counterfactual Evaluation for Explainable AI
• Causal Structure Learning with Recommendation System
• Correcting Exposure Bias for Link Recommendation

Cited by in corpus (3)

• Multi-Cause Deconfounding for Recommender Systems with Latent Confounders
• Decoy Effect In Search Interaction: Understanding User Behavior and Measuring System Vulnerability
• Causality-aware Graph Aggregation Weight Estimator for Popularity Debiasing in Top-K Recommendation