Wei Chen
Ruobing Xie
Skip Abstract Section
Abstract
The emergence of Graph Neural Networks (GNNs) has greatly advanced the development of recommendation systems. Recently, many researchers have leveraged GNN-based models to learn fair representations for users and items. However, current GNN-based models suffer from biased user–item interaction data, which negatively impacts recommendation fairness. Although there have been several studies employing adversarial learning to mitigate this issue in recommendation systems, they mostly focus on modifying the model training approach with fairness regularization and neglect direct intervention of biased interaction. In contrast to these models, this article introduces a novel perspective by directly intervening in observed interactions to generate a counterfactual graph (called FairGap) that is not influenced by sensitive node attributes, enabling us to learn fair representations for users and items easily. We design FairGap to answer the key counterfactual question: “Would interactions with an item remain unchanged if a user’s sensitive attributes were concealed?”. We also provide theoretical proofs to show that our learning strategy via the counterfactual graph is unbiased in expectation. Moreover, we propose a fairness-enhancing mechanism to continuously improve user fairness in the graph-based recommendation. Extensive experimental results against state-of-the-art competitors and base models on three real-world datasets validate the effectiveness of our proposed model.
- [1] . 2020. Towards causal VQA: Revealing and reducing spurious correlations by invariant and covariant semantic editing. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 9690–9698.Google Scholar
Cross Ref
- [2] . 2023. TIGTEC: Token importance guided text counterfactuals. arXiv preprint arXiv:2304.12425 (2023).Google Scholar
- [3] . 2017. Multisided fairness for recommendation. arXiv preprint arXiv:1707.00093 (2017).Google Scholar
- [4] . 2020. Fairness in machine learning: A survey. arXiv preprint arXiv:2010.04053 (2020).Google Scholar
- [5] . 2021. Reinforced counterfactual data augmentation for dual sentiment classification. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. 269–278.Google ScholarCross Ref
- [6] . 2022. Fast variational autoencoder with inverted multi-index for collaborative filtering. In Proceedings of the ACM Web Conference 2022. 1944–1954.Google ScholarDigital Library
- [7] . 2016. Deep neural networks for YouTube recommendations. In Proceedings of the 10th ACM Conference on Recommender Systems. 191–198.Google ScholarDigital Library
- [8] . 2023. Counterfactual thinking. In The Palgrave Encyclopedia of the Possible. Springer, 243–250.Google Scholar
- [9] . 2020. Evaluating stochastic rankings with expected exposure. In Proceedings of the 29th ACM International Conference on Information & Knowledge Management. 275–284.Google ScholarDigital Library
- [10] . 2022. Fairness in graph mining: A survey. arXiv preprint arXiv:2204.09888 (2022).Google Scholar
- [11] . 2023. Beyond-accuracy: A review on diversity, serendipity and fairness in recommender systems based on graph neural networks. arXiv preprint arXiv:2310.02294 (2023).Google Scholar
- [12] . 2021. Exploring author gender in book rating and recommendation. User Modeling and User-Adapted Interaction 31, 3 (2021), 377–420.Google ScholarDigital Library
- [13] . 2023. Progress at protein structure prediction, as seen in CASP15. Current Opinion in Structural Biology 80 (2023), 102594.Google ScholarCross Ref
- [14] . 2022. Recommender forest for efficient retrieval. Advances in Neural Information Processing Systems 35 (2022), 38912–38924.Google Scholar
- [15] . 2023. A survey of graph neural networks for recommender systems: Challenges, methods, and directions. ACM Transactions on Recommender Systems 1, 1 (2023), 1–51.Google ScholarDigital Library
- [16] . 2021. Towards long-term fairness in recommendation. In Proceedings of the 14th ACM International Conference on Web Search and Data Mining. 445–453.Google ScholarDigital Library
- [17] . 2023. Towards understanding fairness and its composition in ensemble machine learning. In 2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE). IEEE, 1533–1545.Google ScholarDigital Library
- [18] . 2020. LightGCN: Simplifying and powering graph convolution network for recommendation. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. 639–648.Google ScholarDigital Library
- [19] . 2023. Addressing confounding feature issue for causal recommendation. ACM Transactions on Information Systems 41, 3 (2023), 1–23.Google ScholarDigital Library
- [20] . 2011. Kullback-Leibler divergence. In International Encyclopedia of Statistical Science. Springer, 720–722.Google ScholarCross Ref
- [21] . 2021. Highly accurate protein structure prediction with AlphaFold. Nature 596, 7873 (2021), 583–589.Google ScholarCross Ref
- [22] . 2023. The homophily principle in social network analysis: A survey. Multimedia Tools and Applications 82, 6 (2023), 8811–8854.Google ScholarDigital Library
- [23] . 2015. Adam: A method for stochastic optimization, 2014. arXiv preprint arxiv:1412.6980 (2015).Google Scholar
- [24] . 2009. Matrix factorization techniques for recommender systems. Computer 42, 8 (2009), 30–37.Google ScholarDigital Library
- [25] . 2010. Randomization tests for distinguishing social influence and homophily effects. In Proceedings of the 19th International Conference on World Wide Web. 601–610.Google ScholarDigital Library
- [26] . 2019. Algorithmic bias? An empirical study of apparent gender-based discrimination in the display of STEM career ads. Management Science 65, 7 (2019), 2966–2981.Google ScholarDigital Library
- [27] . 2023. COCO-counterfactuals: Automatically constructed counterfactual examples for image-text pairs. arXiv preprint arXiv:2309.14356 (2023).Google Scholar
- [28] . 2021. User-oriented fairness in recommendation. In Proceedings of the Web Conference 2021. 624–632.Google ScholarDigital Library
- [29] . 2022. Fairness in recommendation: A survey. arXiv preprint arXiv:2205.13619 (2022).Google Scholar
- [30] . 2021. Towards personalized fairness based on causal notion. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. 1054–1063.Google ScholarDigital Library
- [31] . 2020. LightRec: A memory and search-efficient recommender system. In Proceedings of The Web Conference 2020. 695–705.Google ScholarDigital Library
- [32] . 2023. Learning fair graph representations via automated data augmentations. In The 11th International Conference on Learning Representations.Google Scholar
- [33] . 2022. Mitigating sensitive data exposure with adversarial learning for fairness recommendation systems. Neural Computing and Applications 34, 20 (2022), 18097–18111.Google ScholarDigital Library
- [34] . 2022. Dual constraints and adversarial learning for fair recommenders. Knowledge-Based Systems 239 (2022), 108058.Google ScholarDigital Library
- [35] . 2022. FairEdit: Preserving fairness in graph neural networks through greedy graph editing. arXiv preprint arXiv:2201.03681 (2022).Google Scholar
- [36] . 2023. A survey on causal inference for recommendation. arXiv preprint arXiv:2303.11666 (2023).Google Scholar
- [37] . 2022. Learning fair node representations with graph counterfactual fairness. In Proceedings of the 15th ACM International Conference on Web Search and Data Mining. 695–703.Google ScholarDigital Library
- [38] . 2021. UltraGCN: Ultra simplification of graph convolutional networks for recommendation. In Proceedings of the 30th ACM International Conference on Information & Knowledge Management. 1253–1262.Google ScholarDigital Library
- [39] . 2006. Bayes’ theorem and its applications in animal behaviour. Oikos 112, 2 (2006), 243–251.Google ScholarCross Ref
- [40] . 2001. Birds of a feather: Homophily in social networks. Annual Review of Sociology (2001), 415–444.Google ScholarCross Ref
- [41] . 2021. A survey on bias and fairness in machine learning. ACM Computing Surveys (CSUR) 54, 6 (2021), 1–35.Google ScholarDigital Library
- [42] . 1997. The Jensen-Shannon divergence. Journal of the Franklin Institute 334, 2 (1997), 307–318.Google ScholarCross Ref
- [43] . 2022. CPFair: Personalized consumer and producer fairness re-ranking for recommender systems. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. 770–779.Google ScholarDigital Library
- [44] . 2023. The rise of social media. Our World in Data (2023).Google Scholar
- [45] . 2010. Causal inference. In Proceedings of Workshop on Causality: Objectives and Assessment at (NIPS’08), 39–58.Google Scholar
- [46] . 2011. Scikit-learn: Machine learning in Python. The Journal of Machine Learning Research 12 (2011), 2825–2830.Google ScholarDigital Library
- [47] . 2022. A review on fairness in machine learning. ACM Computing Surveys (CSUR) 55, 3 (2022), 1–44.Google ScholarDigital Library
- [48] . 2014. Attributed graph models: Modeling network structure with correlated attributes. In Proceedings of the 23rd International Conference on World Wide Web. 831–842.Google Scholar
- [49] . 2018. DeepInf: Social influence prediction with deep learning. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2110–2119.Google ScholarDigital Library
- [50] Tahleen Rahman, Bartlomiej Surma, Michael Backes, and Yang Zhang. 2019. Fairwalk: Towards fair graph embedding. In Proceedings of the 28th International Joint Conference on Artificial Intelligence, 3289–3295.Google Scholar
- [51] . 2012. BPR: Bayesian personalized ranking from implicit feedback. arXiv preprint arXiv:1205.2618 (2012).Google Scholar
- [52] . 1997. Counterfactual thinking. Psychological Bulletin 121, 1 (1997), 133.Google ScholarCross Ref
- [53] . 2022. Strategic best response fairness in fair machine learning. In Proceedings of the 2022 AAAI/ACM Conference on AI, Ethics, and Society. 664–664.Google ScholarDigital Library
- [54] . 2021. FairDrop: Biased edge dropout for enhancing fairness in graph representation learning. IEEE Transactions on Artificial Intelligence 3, 3 (2021), 344–354.Google ScholarCross Ref
- [55] . 2020. Robustness to spurious correlations via human annotations. In International Conference on Machine Learning. PMLR, 9109–9119.Google Scholar
- [56] . 2021. Counterfactual explainable recommendation. In Proceedings of the 30th ACM International Conference on Information & Knowledge Management. 1784–1793.Google ScholarDigital Library
- [57] . 2022. Learning to denoise unreliable interactions for graph collaborative filtering. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. 122–132.Google ScholarDigital Library
- [58] . 2022. Fast online ranking with fairness of exposure. In 2022 ACM Conference on Fairness, Accountability, and Transparency. 2157–2167.Google ScholarDigital Library
- [59] . 2022. Counterfactual cycle-consistent learning for instruction following and generation in vision-language navigation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 15471–15481.Google ScholarCross Ref
- [60] . 2022. Unbiased graph embedding with biased graph observations. In Proceedings of the ACM Web Conference 2022. 1423–1433.Google ScholarDigital Library
- [61] . 2021. Clicks can be cheating: Counterfactual recommendation for mitigating clickbait issue. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. 1288–1297.Google ScholarDigital Library
- [62] . 2019. Neural graph collaborative filtering. In Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. 165–174.Google ScholarDigital Library
- [63] . 2020. Causal inference for recommender systems. In Proceedings of the 14th ACM Conference on Recommender Systems. 426–431.Google ScholarDigital Library
- [64] . 2022. A survey on the fairness of recommender systems. Journal of the ACM (JACM) (2022).Google Scholar
- [65] . 2023. A survey on the fairness of recommender systems. ACM Transactions on Information Systems 41, 3 (2023), 1–43.Google ScholarDigital Library
- [66] . 2021. Counterfactual data-augmented sequential recommendation. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. 347–356.Google ScholarDigital Library
- [67] . 2023. Towards fair machine learning software: understanding and addressing model bias through counterfactual thinking. arXiv preprint arXiv:2302.08018 (2023).Google Scholar
- [68] . 2021. Model-agnostic counterfactual reasoning for eliminating popularity bias in recommender system. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. 1791–1800.Google ScholarDigital Library
- [69] . 2023. Influence-driven data poisoning for robust recommender systems. IEEE Transactions on Pattern Analysis and Machine Intelligence (2023).Google ScholarDigital Library
- [70] . 2022. Joint multisided exposure fairness for recommendation. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. 703–714.Google ScholarDigital Library
- [71] . 2021. Learning fair representations for recommendation: A graph-based perspective. In Proceedings of the Web Conference 2021. 2198–2208.Google ScholarDigital Library
- [72] . 2022. A survey on accuracy-oriented neural recommendation: From collaborative filtering to information-rich recommendation. IEEE Transactions on Knowledge and Data Engineering (2022).Google ScholarDigital Library
- [73] . 2022. Graph neural networks in recommender systems: A survey. Comput. Surveys 55, 5 (2022), 1–37.Google ScholarDigital Library
- [74] . 2022. Selective fairness in recommendation via prompts. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2657–2662.Google ScholarDigital Library
- [75] . 2017. Fairness-aware group recommendation with pareto-efficiency. In Proceedings of the 11th ACM Conference on Recommender Systems. 107–115.Google ScholarDigital Library
- [76] . 2021. Counterfactual review-based recommendation. In Proceedings of the 30th ACM International Conference on Information & Knowledge Management. 2231–2240.Google ScholarDigital Library
- [77] . 2021. A survey on causal inference. ACM Transactions on Knowledge Discovery from Data (TKDD) 15, 5 (2021), 1–46.Google ScholarDigital Library
- [78] . 2018. Graph convolutional neural networks for web-scale recommender systems. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 974–983.Google ScholarDigital Library
- [79] . 2022. Towards counterfactual image manipulation via clip. In Proceedings of the 30th ACM International Conference on Multimedia. 3637–3645.Google ScholarDigital Library
- [80] . 2019. Fairness constraints: A flexible approach for fair classification. The Journal of Machine Learning Research 20, 1 (2019), 2737–2778.Google ScholarDigital Library
- [81] . 2023. Towards fairness-aware adversarial network pruning. In Proceedings of the IEEE/CVF International Conference on Computer Vision. 5168–5177.Google ScholarCross Ref
- [82] . 2022. Investigating accuracy-novelty performance for graph-based collaborative filtering. arXiv preprint arXiv:2204.12326 (2022).Google Scholar
Index Terms
- FairGap: Fairness-Aware Recommendation via Generating Counterfactual Graph
Recommendations
Counterfactual Review-based Recommendation
CIKM '21: Proceedings of the 30th ACM International Conference on Information & Knowledge ManagementIncorporating review information into the recommender system has been demonstrated to be an effective method for boosting the recommendation performance. Previous research mainly focus on designing advanced architectures to better profile the users and ...
Counterfactual Explanation for Fairness in Recommendation
Fairness-aware recommendation alleviates discrimination issues to build trustworthy recommendation systems. Explaining the causes of unfair recommendations is critical, as it promotes fairness diagnostics, and thus secures users’ trust in recommendation ...
Using a trust network to improve top-N recommendation
RecSys '09: Proceedings of the third ACM conference on Recommender systemsTop-N item recommendation is one of the important tasks of recommenders. Collaborative filtering is the most popular approach to building recommender systems which can predict ratings for a given user and item. Collaborative filtering can be extended ...
Comments