mlsanityData to 7 Oct 2026

CDR: Conservative Doubly Robust Learning for Debiased Recommendation

arXiv 2308.08461 ↗First posted 13 Aug 2023cs.IR cs.LGcited by 2

Not on the map. Why?

Abstract (arXiv)

In recommendation systems (RS), user behavior data is observational rather than experimental, resulting in widespread bias in the data. Consequently, tackling bias has emerged as a major challenge in the field of recommendation systems. Recently, Doubly Robust Learning (DR) has gained significant attention due to its remarkable performance and robust properties. However, our experimental findings indicate that existing DR methods are severely impacted by the presence of so-called Poisonous Imputation, where the imputation significantly deviates from the truth and becomes counterproductive. To address this issue, this work proposes Conservative Doubly Robust strategy (CDR) which filters imputations by scrutinizing their mean and variance. Theoretical analyses show that CDR offers reduced variance and improved tail bounds.In addition, our experimental investigations illustrate that CDR significantly enhances performance and can indeed reduce the frequency of poisonous imputation.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 31 of its 56 references that are in the map. First posted on arXiv 13 Aug 2023.

Builds on

31 of 56 references.

References · 31

  1. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  2. 2022Diffusion Models in Vision: A Surveycited by 1,008
  3. 2017Generative Adversarial Networks: An Overviewcited by 588
  4. 2015Variational Dropout and the Local Reparameterization Trickcited by 489
  5. 2017Stochastic Gradient Descent as Approximate Bayesian Inferencecited by 298
  6. 2017Variational Dropout Sparsifies Deep Neural Networkscited by 265
  7. 2016Recommendations as Treatments: Debiasing Learning and Evaluationcited by 246
  8. 2020Bias and Debias in Recommender System: A Survey and Future Directionscited by 239
  9. 2017Concrete Dropoutcited by 227
  10. 2018Explainable Recommendation: A Survey and New Perspectivescited by 210
  11. 2017Multiplicative Normalizing Flows for Variational Bayesian Neural Networkscited by 157
  12. 2019A General Framework for Uncertainty Estimation in Deep Learningcited by 119
19 more
  1. 2021Causal Intervention for Leveraging Popularity Bias in Recommendationcited by 116
  2. 2020Model-Agnostic Counterfactual Reasoning for Eliminating Popularity Bias in Recommender Systemcited by 94
  3. 2022A Comprehensive Survey on Deep Clustering: Taxonomy, Challenges, and Future Directionscited by 73
  4. 2022KuaiRand: An Unbiased Sequential Recommendation Dataset with Randomly Exposed Videoscited by 69
  5. 2021AutoDebias: Learning to Debias for Recommendationcited by 68
  6. 2022CIRS: Bursting Filter Bubbles by Counterfactual Interactive Recommender Systemcited by 38
  7. 2023Alleviating Matthew Effect of Offline Reinforcement Learning in Interactive Recommendationcited by 38
  8. 2020Information Theoretic Counterfactual Learning from Missing-Not-At-Random Feedbackcited by 30
  9. 2021Enhanced Doubly Robust Learning for Debiasing Post-click Conversion Rate Estimationcited by 28
  10. 2022On the Opportunity of Causal Learning in Recommendation Systems: Foundation, Estimation, Prediction and Challengescited by 28
  11. 2022StableDR: Stabilized Doubly Robust Learning for Recommendation on Data Missing Not at Randomcited by 24
  12. 2018Deep reinforcement learning for search, recommendation, and online advertising: a surveycited by 23
  13. 2020Multi-sided Exposure Bias in Recommendationcited by 23
  14. 2021Popularity Bias Is Not Always Evil: Disentangling Benign and Harmful Bias for Recommendationcited by 23
  15. 2023Adap-τ: Adaptively Modulating Embedding Magnitude for Recommendationcited by 19
  16. 2022TDR-CL: Targeted Doubly Robust Collaborative Learning for Debiased Recommendationscited by 18
  17. 2022A Generalized Doubly Robust Learning Framework for Debiasing Post-Click Conversion Rate Predictioncited by 18
  18. 2020Fast Adaptively Weighted Matrix Factorization for Recommendation with Implicit Feedbackcited by 17
  19. 2023CPDG: A Contrastive Pre-Training Method for Dynamic Graph Neural Networkscited by 6

Built on by

Cited by year

Citing papers per year: 2024: 1; 2025: 1.

2 papers.

  1. 2024Debiased Recommendation with Noisy Feedbackcited by 8
  2. 2025BPL: Bias-adaptive Preference Distillation Learning for Recommender Systemcited by 0