Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
167 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Be Aware of the Neighborhood Effect: Modeling Selection Bias under Interference (2404.19620v1)

Published 30 Apr 2024 in cs.LG, cs.IR, and stat.ML

Abstract: Selection bias in recommender system arises from the recommendation process of system filtering and the interactive process of user selection. Many previous studies have focused on addressing selection bias to achieve unbiased learning of the prediction model, but ignore the fact that potential outcomes for a given user-item pair may vary with the treatments assigned to other user-item pairs, named neighborhood effect. To fill the gap, this paper formally formulates the neighborhood effect as an interference problem from the perspective of causal inference and introduces a treatment representation to capture the neighborhood effect. On this basis, we propose a novel ideal loss that can be used to deal with selection bias in the presence of neighborhood effect. We further develop two new estimators for estimating the proposed ideal loss. We theoretically establish the connection between the proposed and previous debiasing methods ignoring the neighborhood effect, showing that the proposed methods can achieve unbiased learning when both selection bias and neighborhood effect are present, while the existing methods are biased. Extensive semi-synthetic and real-world experiments are conducted to demonstrate the effectiveness of the proposed methods.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (70)
  1. Estimating average causal effects under general interference, with application to a social network experiment. The Annals of Applied Statistics, 11:1912–1947, 2017.
  2. Prompt-based distribution alignment for unsupervised domain adaptation. In AAAI, 2024.
  3. Autodebias: Learning to debias for recommendation. In SIGIR, 2021a.
  4. Bias and debias in recommender system: A survey and future directions. ACM Transactions on Information Systems, 41(3):1–39, 2023.
  5. Adapting interactional observation embedding for counterfactual learning to rank. In SIGIR, 2021b.
  6. A generalized doubly robust learning framework for debiasing post-click conversion rate prediction. In KDD, 2022.
  7. Addressing unmeasured confounder for recommendation with sensitivity analysis. In KDD, 2022.
  8. Local Polynomial Modelling and Its Applications. Chapman and Hall/CRC, 1996.
  9. Evidence of treatment spillovers within markets. Review of Economics and Statistics, 96:812–823, 2014.
  10. Identification and estimation of treatment and interference effects in observational studies on networks. Journal of the American Statistical Association, 116:901–918, 2021.
  11. KuaiRec: A fully-observed dataset and insights for evaluating recommender systems. In CIKM, 2022.
  12. Enhanced doubly robust learning for debiasing post-click conversion rate estimation. In SIGIR, 2021.
  13. Probabilistic matrix factorization with non-random missing data. In ICML, 2014.
  14. valuating kindergarten retention policy: A case study of causal inference for multilevel observational data. Journal of the American Statistical Association, 101:901–910, 2006.
  15. It is different when items are older: Debiasing recommendations when selection bias and user preferences are dynamic. In WSDM, 2022.
  16. Pareto invariant representation learning for multimedia recommendation. In ACM-MM, 2023.
  17. Toward causal inference with interference. Journal of the American Statistical Association, 103:832–842, 2008.
  18. Nonparametric and Semiparametric Models. Springer Series in Statistics, 2004.
  19. Causal Inference For Statistics Social and Biomedical Science. Cambridge University Press, 2015.
  20. Learning representations for counterfactual inference. In ICML, 2016.
  21. Matrix factorization techniques for recommender systems. Computer, 42(8):30–37, 2009.
  22. Multiple robust learning for recommendation. In AAAI, 2023a.
  23. TDR-CL: Targeted doubly robust collaborative learning for debiased recommendations. In ICLR, 2023b.
  24. Balancing unobserved confounding with a few unbiased ratings in debiased recommendations. In WWW, 2023c.
  25. Propensity matters: Measuring and enhancing balancing for recommendation. In ICML, 2023d.
  26. StableDR: Stabilized doubly robust learning for recommendation on data missing not at random. In ICLR, 2023e.
  27. Removing hidden confounding in recommendation: A unified multi-task learning approach. In NeurIPS, 2023f.
  28. Debiased collaborative filtering with kernel-based causal balancing. In ICLR, 2024.
  29. Qi Li and Jeffrey Scott Racine. Nonparametric econometrics: theory and practice. Princeton University Press, 2023.
  30. Mitigating confounding bias in recommendation via information bottleneck. In RecSys, 2021.
  31. A survey on causal inference for recommendation. The Innovation, 2024.
  32. Duet: A tuning-free device-cloud collaborative parameters generation framework for efficient device model generalization. In WWW, 2023.
  33. Intelligent model update strategy for sequential recommendation. In WWW, 2024.
  34. Entire space multi-task model: An effective approach for estimating post-click conversion rate. In SIGIR, 2018.
  35. Collaborative prediction and ranking with non-random missing data. In RecSys, 2009.
  36. Foundations of Machine Learning. MIT Press, 2018.
  37. Jerzy Splawa Neyman. On the application of probability theory to agricultural experiments. essay on principles. section 9. Statistical Science, 5:465–472, 1990.
  38. Causal diagrams for interference. Statistical Science, 29:559–578, 2014.
  39. Vaccines, contagion, and social networks. The Annals of Applied Statistics, 11:919–948, 2017.
  40. Donald B. Rubin. Estimating causal effects of treatments in randomized and nonrandomized studies. Journal of Educational Psychology, 66:688–701, 1974.
  41. Donald B. Rubin. Discussion of randomization analysis of experimental data in the fisher randomization test by basu. Journal of the American Statistical Association, 75:591–593, 1980.
  42. Yuta Saito. Asymmetric tri-training for debiasing missing-not-at-random explicit feedback. In SIGIR, 2020a.
  43. Yuta Saito. Doubly robust estimator for ranking metrics with post-click conversions. In RecSys, 2020b.
  44. Towards resolving propensity contradiction in offline recommender learning. In IJCAI, 2022.
  45. Unbiased recommender learning from missing-not-at-random implicit feedback. In WSDM, 2020.
  46. Recommendations as treatments: Debiasing learning and evaluation. In ICML, 2016.
  47. Understanding Machine Learning: From Theory to Algorithms. Cambridge University Press, 2014.
  48. Michael E Sobel. What do randomized studies of housing mobility demonstrate? causal inference in the face of interference. Journal of the American Statistical Association, 101:1398–1407, 2006.
  49. Harald Steck. Training and testing of recommender systems on data missing not at random. In KDD, 2010.
  50. The self-normalized estimator for counterfactual learning. In NeurIPS, 2015.
  51. Average treatment effects in the presence of unknown interference. Annals of Statistics, 49:673–701, 2021.
  52. Eric J. Tchetgen Tchetgen and Tyler J. VanderWeele. On causal inference in the presence of interference. Statistical Methods in Medical Research, 21:55–75, 2012.
  53. Auto-g-computation of causal effects on a network. Journal of the American Statistical Association, 116:833–844, 2021.
  54. ESCM2: Entire space counterfactual multi-task model for post-click conversion rate estimation. In SIGIR, 2022a.
  55. Optimal transport for treatment effect estimation. In NeurIPS, 2023a.
  56. Estimating individualized causal effect with confounded instruments. In KDD, 2022b.
  57. Treatment effect estimation with adjustment feature selection. In KDD, 2023b.
  58. Out-of-distribution generalization with causal feature separation. IEEE Transactions on Knowledge and Data Engineering, 36(4):1758–1772, 2024.
  59. CounterCLR: Counterfactual contrastive learning with non-random missing data in recommendation. In ICDM, 2023c.
  60. Causal recommendation: Progresses and future directions. In SIGIR, 2023d.
  61. Doubly robust joint learning for recommendation on data missing not at random. In ICML, 2019.
  62. Combating selection biases in recommender systems with a few unbiased ratings. In WSDM, 2021.
  63. Information theoretic counterfactual learning from missing-not-at-random feedback. In NeurIPS, 2020.
  64. On the opportunity of causal learning in recommendation systems: Foundation, estimation, prediction and challenges. In IJCAI, 2022.
  65. Propensity score regression for causal inference with treatment heterogeneity. Statistica Sinica, 34:747–769, 2024.
  66. Map: Towards balanced generalization of iid and ood through model-agnostic adapters. In ICCV, 2023.
  67. MetaCoCo: A new few-shot classification benchmark with spurious correlation. In ICLR, 2024.
  68. Large-scale causal approaches to debiasing post-click conversion rate estimation with multi-task learning. In WWW, 2020.
  69. Disentangling user interest and conformity for recommendation with causal embedding. In WWW, 2021.
  70. Factual observation based heterogeneity learning for counterfactual prediction. In CLeaR, 2023.
Citations (2)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets