mlsanityData to 14 Oct 2026

Addressing Correlated Latent Exogenous Variables in Debiased Recommender Systems

arXiv 2506.07517 ↗First posted 9 Jun 2025cs.LG cs.IRcited by 0

Not on the map. Why?

Abstract (arXiv)

Recommendation systems (RS) aim to provide personalized content, but they face a challenge in unbiased learning due to selection bias, where users only interact with items they prefer. This bias leads to a distorted representation of user preferences, which hinders the accuracy and fairness of recommendations. To address the issue, various methods such as error imputation based, inverse propensity scoring, and doubly robust techniques have been developed. Despite the progress, from the structural causal model perspective, previous debiasing methods in RS assume the independence of the exogenous variables. In this paper, we release this assumption and propose a learning algorithm based on likelihood maximization to learn a prediction model. We first discuss the correlation and difference between unmeasured confounding and our scenario, then we propose a unified method that effectively handles latent exogenous variables. Specifically, our method models the data generation process with latent exogenous variables under mild normality assumptions. We then develop a Monte Carlo algorithm to numerically estimate the likelihood function. Extensive experiments on synthetic datasets and three real-world datasets demonstrate the effectiveness of our proposed method. The code is at https://github.com/WallaceSUI/kdd25-background-variable.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 22 of its 56 references that are in the map. First posted on arXiv 9 Jun 2025.

Builds on

22 of 56 references.

References · 22

  1. 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,323
  2. 2017Neural Collaborative Filteringcited by 1,111
  3. 2016Recommendations as Treatments: Debiasing Learning and Evaluationcited by 248
  4. 2022KuaiRec: A Fully-observed Dataset and Insights for Evaluating Recommender Systemscited by 96
  5. 2021AutoDebias: Learning to Debias for Recommendationcited by 68
  6. 2023Causal normalizing flows: from theory to practicecited by 51
  7. 2022ESCM2: Entire Space Counterfactual Multi-Task Model for Post-Click Conversion Rate Estimationcited by 47
  8. 2019Large-scale Causal Approaches to Debiasing Post-click Conversion Rate Estimation with Multi-task Learningcited by 39
  9. 2022Indeterminacy in Generative Models: Characterization and Strong Identifiabilitycited by 33
  10. 2020Information Theoretic Counterfactual Learning from Missing-Not-At-Random Feedbackcited by 30
  11. 2019Asymmetric Tri-training for Debiasing Missing-Not-At-Random Explicit Feedbackcited by 28
  12. 2021Enhanced Doubly Robust Learning for Debiasing Post-click Conversion Rate Estimationcited by 28
10 more
  1. 2022On the Opportunity of Causal Learning in Recommendation Systems: Foundation, Estimation, Prediction and Challengescited by 28
  2. 2022StableDR: Stabilized Doubly Robust Learning for Recommendation on Data Missing Not at Randomcited by 24
  3. 2022TDR-CL: Targeted Doubly Robust Collaborative Learning for Debiased Recommendationscited by 18
  4. 2022A Generalized Doubly Robust Learning Framework for Debiasing Post-Click Conversion Rate Predictioncited by 18
  5. 2023Balancing Unobserved Confounding with a Few Unbiased Ratings in Debiased Recommendationscited by 18
  6. 2022Entire Space Counterfactual Learning for Reliable Content Recommendationscited by 13
  7. 2024Doubly Calibrated Estimator for Recommendation on Data Missing Not At Randomcited by 7
  8. 2019Towards Resolving Propensity Contradiction in Offline Recommender Learningcited by 6
  9. 2024Debiased Collaborative Filtering with Kernel-Based Causal Balancingcited by 6
  10. 2024Fine-Grained Dynamic Framework for Bias-Variance Joint Optimization on Data Missing Not at Randomcited by 4

Built on by

Not cited yet.