The Importance of Causality in Decision Making: A Perspective on Recommender Systems

Abstract

Causality is receiving increasing attention in the Recommendation Systems (RSs) community, which has realised that RSs could greatly benefit from causality to transform accurate predictions into effective and explainable decisions. Indeed, the RS literature has repeatedly highlighted that, in real-world scenarios, recommendation algorithms suffer many types of biases since assumptions ensuring unbiasedness are likely not met. In this discussion paper, we formulate the RS problem in terms of causality, using potential outcomes and structural causal models, by giving formal definitions of the causal quantities to be estimated and a general causal graph to serve as a reference to foster future research and development.

Accepted at the CONSEQUENCES '24 workshop, co-located with ACM RecSys '24

References in corpus (32)

• Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learning
• Reinforcement Knowledge Graph Reasoning for Explainable Recommendation
• Causal Intervention for Leveraging Popularity Bias in Recommendation
• Are We Really Making Much Progress? A Worrying Analysis of Recent Neural Recommendation Approaches
• Solving the Cold-Start Problem in Recommender Systems with Social Tags
• Causal Embeddings for Recommendation
• Collaborative Filtering and the Missing at Random Assumption
• Clicks can be Cheating: Counterfactual Recommendation for Mitigating Clickbait Issue
• AutoDebias: Learning to Debias for Recommendation
• Deconfounded Recommendation for Alleviating Bias Amplification
• Estimation of causal effects with multiple treatments: a review and new ideas
• Offline A/B testing for Recommender Systems
• A Survey on Causal Discovery: Theory and Practice
• The Unfairness of Popularity Bias in Recommendation
• Pearl's Calculus of Intervention Is Complete
• Correcting for Selection Bias in Learning-to-rank Systems
• Estimating Position Bias without Intrusive Interventions
• Estimating the Causal Impact of Recommendation Systems from Observational Data
• Unbiased Learning for the Causal Effect of Recommendation
• Cascade Model-based Propensity Estimation for Counterfactual Learning to Rank
• Latent Contextual Bandits and their Application to Personalized Recommendations for New Users
• Multi-sided Exposure Bias in Recommendation
• Causal Inference in Recommender Systems: A Survey and Future Directions
• Decision Making of Maximizers and Satisficers Based on Collaborative Explanations
• Popularity Bias Is Not Always Evil: Disentangling Benign and Harmful Bias for Recommendation
• Causal Disentanglement for Semantics-Aware Intent Learning in Recommendation
• A Study of Position Bias in Digital Library Recommender Systems
• Large-scale Interactive Recommendation with Tree-structured Policy Gradient
• Reinforcement learning based recommender systems: A survey
• Causal Markov Decision Processes: Learning Good Interventions Efficiently
• Evaluating Digital Agriculture Recommendations with Causal Inference
• Causal Structure Learning with Recommendation System