Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
119 tokens/sec
GPT-4o
56 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Causal Learning for Trustworthy Recommender Systems: A Survey (2402.08241v1)

Published 13 Feb 2024 in cs.IR

Abstract: Recommender Systems (RS) have significantly advanced online content discovery and personalized decision-making. However, emerging vulnerabilities in RS have catalyzed a paradigm shift towards Trustworthy RS (TRS). Despite numerous progress on TRS, most of them focus on data correlations while overlooking the fundamental causal nature in recommendation. This drawback hinders TRS from identifying the cause in addressing trustworthiness issues, leading to limited fairness, robustness, and explainability. To bridge this gap, causal learning emerges as a class of promising methods to augment TRS. These methods, grounded in reliable causality, excel in mitigating various biases and noises while offering insightful explanations for TRS. However, there lacks a timely survey in this vibrant area. This paper creates an overview of TRS from the perspective of causal learning. We begin by presenting the advantages and common procedures of Causality-oriented TRS (CTRS). Then, we identify potential trustworthiness challenges at each stage and link them to viable causal solutions, followed by a classification of CTRS methods. Finally, we discuss several future directions for advancing this field.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (61)
  1. Controlling popularity bias in learning-to-rank recommendation. In RecSys, pages 42–46, 2017.
  2. Personalized fashion recommendation with visual explanations based on multimodal attention network: Towards visually explainable recommendation. In SIGIR, pages 765–774, 2019.
  3. GREASE: generate factual and counterfactual explanations for gnn-based recommendations. CoRR, abs/2208.04222, 2022.
  4. A comprehensive survey on trustworthy recommender systems. CoRR, abs/2209.10117, 2022.
  5. Causal inference in recommender systems: A survey and future directions. CoRR, abs/2208.12397, 2022.
  6. CIRS: bursting filter bubbles by counterfactual interactive recommender system. TOIS, 42(1):14:1–14:27, 2024.
  7. A survey on trustworthy recommender systems. CoRR, abs/2207.12515, 2022.
  8. Causer: Causal session-based recommendations for handling popularity bias. In CIKM, pages 3048–3052, 2021.
  9. Addressing confounding feature issue for causal recommendation. TOIS, 41(3):53:1–53:23, 2023.
  10. Probabilistic matrix factorization with non-random missing data. In ICML, volume 32, pages 1512–1520, 2014.
  11. Efficient estimation of average treatment effects using the estimated propensity score. Econometrica, 71(4):1161–1189, 2003.
  12. Constraint-based causal discovery: Conflict resolution with answer set programming. In UAI, pages 340–349, 2014.
  13. Evaluating the accuracy of implicit feedback from clicks and query reformulations in web search. TOIS, 25(2):7, 2007.
  14. Unbiased learning-to-rank with biased feedback. In IJCAI, pages 5284–5288, 2018.
  15. Towards personalized fairness based on causal notion. In SIGIR, pages 1054–1063, 2021.
  16. Balancing unobserved confounding with a few unbiased ratings in debiased recommendations. In The Web Conference, pages 1305–1313, 2023.
  17. Propensity matters: Measuring and enhancing balancing for recommendation. In ICML, volume 202, pages 20182–20194, 2023.
  18. Fairness in recommendation: Foundations, methods, and applications. TIST, 14(5):95:1–95:48, 2023.
  19. Improving graph collaborative filtering with neighborhood-enriched contrastive learning. In The Web Conference, pages 2320–2329, 2022.
  20. Improving causal discovery by optimal bayesian network learning. In AAAI, pages 8741–8748, 2021.
  21. A survey on causal inference for recommendation. CoRR, abs/2303.11666, 2023.
  22. Memory augmented graph neural networks for sequential recommendation. In AAAI, pages 5045–5052, 2020.
  23. Investigating potential factors associated with gender discrimination in collaborative recommender systems. In FLAIRS, pages 193–196, 2020.
  24. Feedback loop and bias amplification in recommender systems. In CIKM, pages 2145–2148, 2020.
  25. Alleviating spurious correlations in knowledge-aware recommendations through counterfactual generator. In SIGIR, pages 1401–1411, 2022.
  26. Judea Pearl. Causality. Cambridge university press, 2009.
  27. Counterfactual learning and evaluation for recommender systems: Foundations, implementations, and recent advances. In RecSys, pages 828–830, 2021.
  28. Uplift-based evaluation and optimization of recommenders. In RecSys, pages 296–304, 2019.
  29. Recommendations as treatments: Debiasing learning and evaluation. In ICML, volume 48, pages 1670–1679, 2016.
  30. Bert4rec: Sequential recommendation with bidirectional encoder representations from transformer. In CIKM, pages 1441–1450, 2019.
  31. Counterfactual explainable recommendation. In CIKM, pages 1784–1793, 2021.
  32. Adversarial training towards robust multimedia recommender system. IEEE TKDE, 32(5):855–867, 2020.
  33. Tagsplanations: explaining recommendations using tags. In IUI, pages 47–56, 2009.
  34. Doubly robust joint learning for recommendation on data missing not at random. In ICML, volume 97, pages 6638–6647, 2019.
  35. Clicks can be cheating: Counterfactual recommendation for mitigating clickbait issue. In SIGIR, pages 1288–1297, 2021.
  36. Deconfounded recommendation for alleviating bias amplification. In SIGKDD, pages 1717–1725, 2021.
  37. Denoising implicit feedback for recommendation. In WSDM, pages 373–381, 2021.
  38. Learning to augment for casual user recommendation. In The Web Conference, pages 2183–2194, 2022.
  39. User-controllable recommendation against filter bubbles. In SIGIR, pages 1251–1261, 2022.
  40. Trustworthy recommender systems. TIST, 2023.
  41. Model-agnostic counterfactual reasoning for eliminating popularity bias in recommender system. In SIGKDD, pages 1791–1800, 2021.
  42. Learning fair representations for recommendation: A graph-based perspective. In The Web Conference, pages 2198–2208, 2021.
  43. Are big recommendation models fair to cold users? CoRR, abs/2202.13607, 2022.
  44. On the opportunity of causal learning in recommendation systems: Foundation, estimation, prediction and challenges. In IJCAI, pages 5646–5653, 2022.
  45. Reinforcement knowledge graph reasoning for explainable recommendation. In SIGIR, pages 285–294, 2019.
  46. Causal collaborative filtering. In ICTIR, pages 235–245, 2023.
  47. Causal inference for recommendation: Foundations, methods and applications. CoRR, abs/2301.04016, 2023.
  48. Deconfounded causal collaborative filtering. TORS, 1(4):1–25, 2023.
  49. Knowledge graph contrastive learning for recommendation. In SIGIR, pages 1434–1443, 2022.
  50. A feedback shift correction in predicting conversion rates under delayed feedback. In The Web Conference, pages 2740–2746, 2020.
  51. Counterfactual explainable conversational recommendation. TKDE, pages 1–13, 2023.
  52. Yongfeng Zhang and Xu Chen. Explainable recommendation: A survey and new perspectives. FTIR, 14(1):1–101, 2020.
  53. Gcn-based user representation learning for unifying robust recommendation and fraudster detection. In SIGIR, pages 689–698, 2020.
  54. Large-scale causal approaches to debiasing post-click conversion rate estimation with multi-task learning. In The Web Conference, pages 2775–2781, 2020.
  55. Causerec: Counterfactual user sequence synthesis for sequential recommendation. In SIGIR, pages 367–377, 2021.
  56. Counterfactual reward modification for streaming recommendation with delayed feedback. In SIGIR, pages 41–50, 2021.
  57. Causal intervention for leveraging popularity bias in recommendation. In SIGIR, pages 11–20, 2021.
  58. Pipattack: Poisoning federated recommender systems for manipulating item promotion. In WSDM, pages 1415–1423, 2022.
  59. Disentangling user interest and conformity for recommendation with causal embedding. In The Web Conference, pages 2980–2991, 2021.
  60. Mitigating hidden confounding effects for causal recommendation. CoRR, abs/2205.07499, 2022.
  61. Causal inference in recommender systems: A survey of strategies for bias mitigation, explanation, and generalization. CoRR, abs/2301.00910, 2023.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (8)
  1. Jin Li (366 papers)
  2. Shoujin Wang (40 papers)
  3. Qi Zhang (785 papers)
  4. Longbing Cao (85 papers)
  5. Fang Chen (97 papers)
  6. Xiuzhen Zhang (35 papers)
  7. Dietmar Jannach (53 papers)
  8. Charu C. Aggarwal (29 papers)

Summary

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

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