Cross-Dataset Propensity Estimation for Debiasing Recommender Systems

2022/12/22 by Fengyu Li, Li, Fengyu, Sarah Dean +1
Computer Science · #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Machine Learning (cs.LG) #Machine Learning and Data Classification #Methodology (stat.ME) #Recommender Systems and Techniques

paper · pdf · doi:10.48550/arxiv.2212.13892

openalex publication_date 2022/12/22 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Datasets for training recommender systems are often subject to distribution shift induced by users' and recommenders' selection biases. In this paper, we study the impact of selection bias on datasets with different quantization. We then leverage two differently quantized datasets from different source distributions to mitigate distribution shift by applying the inverse probability scoring method from causal inference. Empirically, our approach gains significant performance improvement over single-dataset methods and alternative ways of combining two datasets.

Related

• Addressing Correlated Latent Exogenous Variables in Debiased Recommender Systems
• Recommendations as Treatments: Debiasing Learning and Evaluation
• Debiased Recommendation with User Feature Balancing
• Cross Pairwise Ranking for Unbiased Item Recommendation
• Bounding System-Induced Biases in Recommender Systems with A Randomized Dataset