Abstract
Recently, there has been a growing interest in mitigating the bias effects in recommendations using causal inference. However, Rubin’s potential outcome framework may produce inaccurate estimates in real-world scenarios due to the presence of hidden confounders. In addition, existing works adopting the Pearl causal graph framework tend to focus on specific types of bias (e.g., selection bias, popularity bias, exposure bias) instead of directly mitigating the impact of hidden confounders. Motivated by the aforementioned limitations, in this paper, we formulate the recommendation task as a causal graph with unobserved/unmeasurable confounders. We present a novel causality-based architecture called Multi-behavior Debiased Contrastive Collaborative Filtering (MDCCL) and apply the front-door adjustment for intervention. We leverage a pre-like behavior such as clicking an item (i.e., a behavior occurred before the target behavior such as purchasing) to mitigate the bias effects. Additionally, we design a contrastive loss that also provides a debiasing effect benefiting the recommendation. An empirical study on three real-world datasets validates that our proposed method successfully outperforms nine state-of-the-art baselines. Code and the datasets will be available at https://github.com/queenjocey/MDCCL.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Abdollahpouri, H., Burke, R., Mobasher, B.: Controlling popularity bias in learning-to-rank recommendation. In: RecSys, pp. 42–46 (2017)
Chen, J., et al.: AutoDebias: learning to debias for recommendation. In: SIGIR, pp. 21–30 (2021)
Chen, J., Dong, H., Wang, X., Feng, F., Wang, M., He, X.: Bias and debias in recommender system: a survey and future directions. ACM Trans. Inf. Syst. 41, 1–39 (2020)
Ding, S., et al.: Addressing unmeasured confounder for recommendation with sensitivity analysis. In: KDD, pp. 305–315 (2022)
Gao, C., et al.: KuaiRec: a fully-observed dataset and insights for evaluating recommender systems. In: CIKM (2022)
Gulla, J.A., Zhang, L., Liu, P., Özgöbek, O., Su, X.: The Adressa dataset for news recommendation. In: WI (2017)
Guo, S., et al.: Enhanced doubly robust learning for debiasing post-click conversion rate estimation. In: SIGIR (2021)
Gutmann, M., Hyärinen, A.: Noise-contrastive estimation: a new estimation principle for unnormalized statistical models. In: AISTATS, pp. 297–304. JMLR Workshop and Conference Proceedings (2010)
He, R., McAuley, J.: Ups and downs: modeling the visual evolution of fashion trends with one-class collaborative filtering. In: WWW, pp. 507–517 (2016)
He, X., Deng, K., Wang, X., Li, Y., Zhang, Y., Wang, M.: LightGCN: simplifying and powering graph convolution network for recommendation. In: SIGIR, pp. 639–648 (2020)
He, X., Liao, L., Zhang, H., Nie, L., Hu, X., Chua, T.S.: Neural collaborative filtering. In: WWW, pp. 173–182 (2017)
Hernández-Lobato, J.M., Houlsby, N., Ghahramani, Z.: Probabilistic matrix factorization with non-random missing data. In: ICML (2014)
Kim, Y., Hassan, A., White, R.W., Zitouni, I.: Modeling dwell time to predict click-level satisfaction. In: WSDM, pp. 193–202 (2014)
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: Bengio, Y., LeCun, Y. (eds.) ICLR (2015)
Koren, Y., Bell, R., Volinsky, C.: Matrix factorization techniques for recommender systems. Computer 42(8), 30–37 (2009)
Li, Y., Hu, J., Zhai, C., Chen, Y.: Improving one-class collaborative filtering by incorporating rich user information. In: CIKM (2010)
Liang, D., Krishnan, R.G., Hoffman, M.D., Jebara, T.: Variational autoencoders for collaborative filtering. In: WWW, pp. 689–698 (2018)
Liu, D., et al.: Mitigating confounding bias in recommendation via information bottleneck. In: RecSys (2021)
Liu, D., Lin, C., Zhang, Z., Xiao, Y., Tong, H.: Spiral of silence in recommender systems. In: WSDM, pp. 222–230 (2019)
Ma, X., et al.: Entire space multi-task model: an effective approach for estimating post-click conversion rate. In: SIGIR (2018)
van den Oord, A., Li, Y., Vinyals, O.: Representation learning with contrastive predictive coding. arXiv preprint (2018)
Ovaisi, Z., Ahsan, R., Zhang, Y., Vasilaky, K., Zheleva, E.: Correcting for selection bias in learning-to-rank systems. In: WWW (2020)
Pearl, J., Glymour, M.M., Jewell, N.P.: Causal inference in statistics: a primer (2016)
Rendle, S., Freudenthaler, C., Gantner, Z., Schmidt-Thieme, L.: BPR: Bayesian personalized ranking from implicit feedback. In: UAI, pp. 452–461 (2009)
Saito, Y.: Asymmetric tri-training for debiasing missing-not-at-random explicit feedback. In: SIGIR (2020)
Saito, Y.: Unbiased pairwise learning from biased implicit feedback. In: SIGIR, pp. 5–12 (2020)
Schnabel, T., Swaminathan, A., Singh, A., Chandak, N., Joachims, T.: Recommendations as treatments: debiasing learning and evaluation, pp. 1670–1679. PMLR (2016)
Tao, W., et al. : SMINet: state-aware multi-aspect interests representation network for cold-start users recommendation. In: AAAI (2022)
Tran, T., Sweeney, R., Lee, K.: Adversarial Mahalanobis distance-based attentive song recommender for automatic playlist continuation. In: SIGIR (2019)
Tran, T., You, D., Lee, K.: Quaternion-based self-attentive long short-term user preference encoding for recommendation. In: CIKM, pp. 1455–1464 (2020)
VanderWeele, T.J.: A three-way decomposition of a total effect into direct, indirect, and interactive effects. Epidemiology 24, 224–232 (2013)
Wang, W., Feng, F., He, X., Wang, X., Chua, T.S.: Deconfounded recommendation for alleviating bias amplification. In: KDD, pp. 1717–1725 (2021)
Wang, W., Feng, F., He, X., Zhang, H., Chua, T.S.: Clicks can be cheating: counterfactual recommendation for mitigating clickbait issue. In: SIGIR (2021)
Wang, X., He, X., Wang, M., Feng, F., Chua, T.S.: Neural graph collaborative filtering. In: SIGIR, pp. 165–174 (2019)
Wang, X., Zhang, R., Sun, Y., Qi, J.: Doubly robust joint learning for recommendation on data missing not at random. In: ICML (2019)
Wei, C., Liang, J., Liu, D., Wang, F.: Contrastive graph structure learning via information bottleneck for recommendation. In: NeurIPS, vol. 35, pp. 20407–20420 (2022)
Wei, T., Feng, F., Chen, J., Wu, Z., Yi, J., He, X.: Model-agnostic counterfactual reasoning for eliminating popularity bias in recommender system. In: KDD (2021)
Xin, X., He, X., Zhang, Y., Zhang, Y., Jose, J.: Relational collaborative filtering: modeling multiple item relations for recommendation. In: SIGIR, pp. 125–134 (2019)
Xu, S., Tan, J., Heinecke, S., Li, J., Zhang, Y.: Deconfounded causal collaborative filtering. arXiv arXiv:2110.07122 (2021)
Yang, L., Cui, Y., Xuan, Y., Wang, C., Belongie, S., Estrin, D.: Unbiased offline recommender evaluation for missing-not-at-random implicit feedback. In: RecSys, pp. 279–287 (2018)
You, D., Vo, N., Lee, K., Liu, Q.: Attributed multi-relational attention network for fact-checking URL recommendation. In: CIKM (2019)
Zhang, W., et al.: Large-scale causal approaches to debiasing post-click conversion rate estimation with multi-task learning. In: WWW (2020)
Zhang, Y., et al.: Causal intervention for leveraging popularity bias in recommendation. In: SIGIR (2021)
Zhao, Z., et al.: Popularity bias is not always evil: disentangling benign and harmful bias for recommendation. IEEE Trans. Knowl. Data Eng. 35, 9920–9931 (2022)
Zheng, Y., Gao, C., Li, X., He, X., Li, Y., Jin, D.: Disentangling user interest and conformity for recommendation with causal embedding. In: Proceedings of the Web Conference 2021, pp. 2980–2991 (2021)
Zheng, Y., Gao, C., Li, X., He, X., Li, Y., Jin, D.: Disentangling user interest and conformity for recommendation with causal embedding. In: WWW (2021)
Zhu, X., Zhang, Y., Feng, F., Yang, X., Wang, D., He, X.: Mitigating hidden confounding effects for causal recommendation. arXiv arXiv:2205.07499 (2022)
Zhu, Z., He, Y., Zhang, Y., Caverlee, J.: Unbiased implicit recommendation and propensity estimation via combinational joint learning. In: RecSys (2020)
Zhu, Z., He, Y., Zhao, X., Caverlee, J.: Popularity bias in dynamic recommendation. In: KDD (2021)
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
You, D., Lee, K. (2024). Alleviating Confounding Effects with Contrastive Learning in Recommendation. In: Goharian, N., et al. Advances in Information Retrieval. ECIR 2024. Lecture Notes in Computer Science, vol 14609. Springer, Cham. https://doi.org/10.1007/978-3-031-56060-6_30
Download citation
DOI: https://doi.org/10.1007/978-3-031-56060-6_30
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-56059-0
Online ISBN: 978-3-031-56060-6
eBook Packages: Computer ScienceComputer Science (R0)