{"type":"ScholarlyArticle","title":"Discriminative-Invariant Representation Learning for Unbiased Recommendation","authors":[{"name":"Hang Pan","slug":"hang-pan","given":"Hang","family":"Pan","affiliation":"University of Science and Technology of China"},{"name":"Jiawei Chen","slug":"jiawei-chen","given":"Jiawei","family":"Chen","affiliation":"Zhejiang University"},{"name":"Fuli Feng","slug":"fuli-feng","given":"Fuli","family":"Feng","affiliation":"University of Science and Technology of China; Institute of Dataspace"},{"name":"Wentao Shi","slug":"wentao-shi","given":"Wentao","family":"Shi","affiliation":"University of Science and Technology of China"},{"name":"Junkang Wu","slug":"junkang-wu","given":"Junkang","family":"Wu","affiliation":"University of Science and Technology of China"},{"name":"Xiangnan He","slug":"xiangnan-he","given":"Xiangnan","family":"He","affiliation":"University of Science and Technology of China"}],"abstract":"Selection bias hinders recommendation models from learning unbiased user preference. Recent works empirically reveal that pursuing invariant user and item representation across biased and unbiased data is crucial for counteracting selection bias. However, our theoretical analysis reveals that simply optimizing representation invariance is insufficient for addressing the selection bias — recommendation performance is bounded by both representation invariance and discriminability. Worse still, current invariant representation learning methods in recommendation neglect even hurt the representation discriminability due to data sparsity and label shift. In this light, we propose a new Discriminative-Invariant Representation Learning framework for unbiased recommendation, which incorporates label-conditional clustering and prior-guided contrasting into conventional invariant representation learning to mitigate the impact of data sparsity and label shift, respectively. We conduct extensive experiments on three real-world datasets, validating the rationality and effectiveness of the proposed framework. Code and supplementary materials are available at: https://github.com/HungPaan/DIRL.","doi":"10.24963/ijcai.2023/252","url":"https://ijcai.org/proceedings/2023/252","pdf":"https://ijcai.org/proceedings/2023/0252.pdf","markdown":"https://ijcai.org/proceedings/2023/252.md","bibtex":"https://ijcai.org/proceedings/2023/bibtex/252","volume":{"slug":"2023","title":"Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2023","url":"https://ijcai.org/proceedings/2023/","year":2023,"location":"Macao, SAR","dates":"19-25 August 2023","editors":"Edith Elkind","isbn":"978-1-956792-03-4","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2023"},"section":"Data Mining","track":"Main Track","number":252,"pages":{"first":2270,"last":2278},"keywords":["Data Mining: DM: Recommender systems","AI Ethics, Trust, Fairness: ETF: Bias"],"videos":[],"references":[{"position":1,"raw":"[Acuna et al., 2021] David Acuna, Guojun Zhang, Marc T. Law, and Sanja Fidler. f-domain adversarial learning: The- ory and algorithms. In ICML, pages 66-75, 2021.","title":"f-domain adversarial learning: Theory and algorithms","authors":"David Acuna, Guojun Zhang, Marc T Law, Sanja Fidler","year":"2021","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Ben-David et al., 2010] Shai Ben-David, John Blitzer, Koby Crammer, Alex Kulesza, Fernando Pereira, and Jennifer Wortman Vaughan. A theory of learning from different domains. Mach. Learn., 79:151-175, 2010.","title":"A theory of learning from different domains","authors":"Shai Ben-David, John Blitzer, Koby Crammer, Alex Kulesza, Fernando Pereira, Jennifer Wortman Vaughan","year":"2010","venue":"Mach. Learn","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Chen et al., 2019] Xinyang Chen, Sinan Wang, Mingsheng Long, and Jianmin Wang. Transferability vs. discrim- inability: Batch spectral penalization for adversarial do- main adaptation. In ICML, pages 1081-1090, 2019.","title":"Transferability vs. discriminability: Batch spectral penalization for adversarial domain adaptation","authors":"Xinyang Chen, Sinan Wang, Mingsheng Long, Jianmin Wang","year":"2019","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Chen et al., 2021] Jiawei Chen, Hande Dong, Yang Qiu, Xi- angnan He, Xin Xin, Liang Chen, Guli Lin, and Keping Yang. Autodebias: Learning to debias for recommenda- tion. In SIGIR, pages 21-30, 2021.","title":"Autodebias: Learning to debias for recommendation","authors":"Jiawei Chen, Hande Dong, Yang Qiu, Xiangnan He, Xin Xin, Liang Chen, Guli Lin, Keping Yang","year":"2021","venue":"SIGIR","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Chen et al., 2023] Jiawei Chen, Hande Dong, Xiang Wang, Fuli Feng, Meng Wang, and Xiangnan He. Bias and debias in recommender system: A survey and future directions. TOIS, 41(3), 2023.","title":"Bias and debias in recommender system: A survey and future directions","authors":"Jiawei Chen, Hande Dong, Xiang Wang, Fuli Feng, Meng Wang, Xiangnan He","year":"2023","venue":"TOIS","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Dai et al., 2022] Quanyu Dai, Haoxuan Li, Peng Wu, Zhen- hua Dong, Xiao-Hua Zhou, Rui Zhang, Rui Zhang, and Jie Sun. A generalized doubly robust learning frame- work for debiasing post-click conversion rate prediction. In SIGKDD, pages 252-262, 2022.","title":"A generalized doubly robust learning framework for debiasing post-click conversion rate prediction","authors":"Dai","year":"2022","venue":"SIGKDD","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Deng et al., 2019] Zhijie Deng, Yucen Luo, and Jun Zhu. Cluster alignment with a teacher for unsupervised domain adaptation. In ICCV, pages 9943-9952, 2019.","title":"Cluster alignment with a teacher for unsupervised domain adaptation","authors":"Zhijie Deng, Yucen Luo, Jun Zhu","year":"2019","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Ganin and Lempitsky, 2015] Yaroslav Ganin and Victor S. Lempitsky. Unsupervised domain adaptation by backprop- agation. In ICML, pages 1180-1189, 2015.","title":"Unsupervised domain adaptation by backpropagation","authors":"Yaroslav Ganin, Victor S Lempitsky","year":"2015","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Guo et al., 2021] Siyuan Guo, Lixin Zou, Yiding Liu, Wen- wen Ye, Suqi Cheng, Shuaiqiang Wang, Hechang Chen, Dawei Yin, and Yi Chang. Enhanced doubly robust learn- ing for debiasing post-click conversion rate estimation. In SIGIR, pages 275-284, 2021.","title":"Enhanced doubly robust learning for debiasing post-click conversion rate estimation","authors":"Siyuan Guo, Lixin Zou, Yiding Liu, Wenwen Ye, Suqi Cheng, Shuaiqiang Wang, Hechang Chen, Dawei Yin, Yi Chang","year":"2021","venue":"SIGIR","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Huang et al., 2022] Jin Huang, Harrie Oosterhuis, and Maarten de Rijke. It is different when items are older: Debiasing recommendations when selection bias and user preferences are dynamic. In WSDM, pages 381-389, 2022.","title":"It is different when items are older: Debiasing recommendations when selection bias and user preferences are dynamic","authors":"Jin Huang, Harrie Oosterhuis, Maarten De Rijke","year":"2022","venue":"WSDM","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Imbens and Rubin, 2015] Guido W Imbens and Donald B Rubin. Causal inference in statistics, social, and biomedi- cal sciences. Cambridge University Press, 2015.","title":"Causal inference in statistics, social, and biomedical sciences","authors":"W Guido, Donald B Imbens, Rubin","year":"2015","venue":null,"doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Jiang and Li, 2016] Nan Jiang and Lihong Li. Doubly ro- bust off-policy value evaluation for reinforcement learn- ing. In ICML, pages 652-661, 2016.","title":"Doubly robust off-policy value evaluation for reinforcement learning","authors":"Nan Jiang, Lihong Li","year":"2016","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Jiang et al., 2022] Junguang Jiang, Yang Shu, Jianmin Wang, and Mingsheng Long. Transferability in deep learn- ing: A survey. CoRR, abs/2201.05867, 2022.","title":"Transferability in deep learning: A survey","authors":"Jiang","year":"2022","venue":null,"doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Kang et al., 2019] Guoliang Kang, Lu Jiang, Yi Yang, and Alexander G. Hauptmann. Contrastive adaptation net- work for unsupervised domain adaptation. In CVPR, pages 4893-4902, 2019.","title":"Contrastive adaptation network for unsupervised domain adaptation","authors":"Guoliang Kang, Lu Jiang, Yi Yang, Alexander G Hauptmann","year":"2019","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Koren et al., 2009] Yehuda Koren, Robert M. Bell, and Chris Volinsky. Matrix factorization techniques for rec- ommender systems. Computer, 42:30-37, 2009.","title":"Matrix factorization techniques for recommender systems","authors":"Koren","year":"2009","venue":"Computer","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Kundu et al., 2022] Jogendra Nath Kundu, Akshay R. Kulkarni, Suvaansh Bhambri, Deepesh Mehta, Shreyas Anand Kulkarni, Varun Jampani, and Venkatesh Babu Radhakrishnan. Balancing discrim- inability and transferability for source-free domain adaptation. In ICML, pages 11710-11728, 2022.","title":"Balancing discriminability and transferability for source-free domain adaptation","authors":"Jogendra Nath Kundu, Akshay R Kulkarni, Suvaansh Bhambri, Deepesh Mehta, Anand Shreyas, Varun Kulkarni, Venkatesh Jampani, Babu Radhakrishnan","year":"2022","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Lee et al., 2022] Jae-woong Lee, Seongmin Park, Joonseok Lee, and Jongwuk Lee. Bilateral self-unbiased learning from biased implicit feedback. In SIGIR, pages 29-39, 2022.","title":"Bilateral self-unbiased learning from biased implicit feedback","authors":"Jae-Woong Lee, Seongmin Park, Joonseok Lee, Jongwuk Lee","year":"2022","venue":"SIGIR","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Lee, 2013] Dong-Hyun Lee. Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks. In Workshop on challenges in representation learning, ICML, page 896, 2013.","title":"Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks","authors":"Dong-Hyun Lee","year":"2013","venue":"Workshop on challenges in representation learning","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Li et al., 2020] Bo Li, Yezhen Wang, Tong Che, Shang- hang Zhang, Sicheng Zhao, Pengfei Xu, Wei Zhou, Yoshua Bengio, and Kurt Keutzer. Rethinking dis- tributional matching based domain adaptation. CoRR, abs/2006.13352, 2020.","title":"Rethinking distributional matching based domain adaptation","authors":"Bo Li, Yezhen Wang, Tong Che, Shanghang Zhang, Sicheng Zhao, Pengfei Xu, Wei Zhou, Yoshua Bengio, Kurt Keutzer","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Li et al., 2022a] Haoxuan Li, Quanyu Dai, Yuru Li, Yan Lyu, Zhenhua Dong, Peng Wu, and Xiao-Hua Zhou. Mul- tiple robust learning for recommendation. arXiv preprint arXiv:2207.10796, 2022.","title":"Multiple robust learning for recommendation","authors":"Li","year":"2022","venue":null,"doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Li et al., 2022b] Haoxuan Li, Chunyuan Zheng, Xiao-Hua Zhou, and Peng Wu. Stabilized doubly robust learning for recommendation on data missing not at random. arXiv preprint arXiv:2205.04701, 2022.","title":"Stabilized doubly robust learning for recommendation on data missing not at random","authors":"Li","year":"2022","venue":null,"doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Liu et al., 2020] Dugang Liu, Pengxiang Cheng, Zhenhua Dong, Xiuqiang He, Weike Pan, and Zhong Ming. A gen- eral knowledge distillation framework for counterfactual recommendation via uniform data. In SIGIR, pages 831- 840, 2020.","title":"A general knowledge distillation framework for counterfactual recommendation via uniform data","authors":"Liu","year":"2020","venue":"SIGIR","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Liu et al., 2021a] Dugang Liu, Pengxiang Cheng, Hong Zhu, Zhenhua Dong, Xiuqiang He, Weike Pan, and Zhong Ming. Mitigating confounding bias in recommendation via information bottleneck. In RecSys, pages 351-360, 2021.","title":"Mitigating confounding bias in recommendation via information bottleneck","authors":"Liu","year":"2021","venue":"RecSys","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Liu et al., 2021b] Xiaofeng Liu, Zhenhua Guo, Site Li, Fangxu Xing, Jane You, C.-C. Jay Kuo, Georges El Fakhri, and Jonghye Woo. Adversarial unsupervised domain adap- tation with conditional and label shift: Infer, align and it- erate. In ICCV, pages 10347-10356, 2021.","title":"Adversarial unsupervised domain adaptation with conditional and label shift: Infer, align and iterate","authors":"Liu","year":"2021","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Prabhu et al., 2021] Viraj Prabhu, Shivam Khare, Deeksha Kartik, and Judy Hoffman. SENTRY: selective entropy optimization via committee consistency for unsupervised domain adaptation. In ICCV, pages 8538-8547, 2021.","title":"SENTRY: selective entropy optimization via committee consistency for unsupervised domain adaptation","authors":"Viraj Prabhu, Shivam Khare, Deeksha Kartik, Judy Hoffman","year":"2021","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Saito and Nomura, 2022] Yuta Saito and Masahiro Nomura. Towards resolving propensity contradiction in offline rec- ommender learning. In IJCAI, pages 2211-2217, 2022.","title":"Towards resolving propensity contradiction in offline recommender learning","authors":"Yuta Saito, Masahiro Nomura","year":"2022","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2022/307"},{"position":27,"raw":"[Saito, 2020a] Yuta Saito. Asymmetric tri-training for debi- asing missing-not-at-random explicit feedback. In SIGIR, pages 309-318, 2020.","title":"Asymmetric tri-training for debiasing missing-not-at-random explicit feedback","authors":"Yuta Saito, Saito","year":"2020","venue":"SIGIR","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Saito, 2020b] Yuta Saito. Unbiased pairwise learning from biased implicit feedback. In ICTIR, pages 5-12, 2020.","title":"Unbiased pairwise learning from biased implicit feedback","authors":"Yuta Saito, Saito","year":"2020","venue":"ICTIR","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Schnabel et al., 2016] Tobias Schnabel, Adith Swami- nathan, Ashudeep Singh, Navin Chandak, and Thorsten Joachims. Recommendations as treatments: Debiasing learning and evaluation. In ICML, pages 1670-1679, 2016.","title":"Recommendations as treatments: Debiasing learning and evaluation","authors":"Schnabel","year":"2016","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Swaminathan and Joachims, 2015] Adith Swaminathan and Thorsten Joachims. The self-normalized estimator for counterfactual learning. In NeurIPS, pages 3231-3239, 2015.","title":"The self-normalized estimator for counterfactual learning","authors":"Joachims Swaminathan, Adith Swaminathan, Thorsten Joachims","year":"2015","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":31,"raw":"[Tzeng et al., 2015] Eric Tzeng, Judy Hoffman, Trevor Dar- rell, and Kate Saenko. Simultaneous deep transfer across domains and tasks. In ICCV, pages 4068-4076, 2015.","title":"Simultaneous deep transfer across domains and tasks","authors":"Eric Tzeng, Judy Hoffman, Trevor Darrell, Kate Saenko","year":"2015","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":32,"raw":"[Tzeng et al., 2017] Eric Tzeng, Judy Hoffman, Kate Saenko, and Trevor Darrell. Adversarial discriminative domain adaptation. In CVPR, pages 2962-2971, 2017.","title":"Adversarial discriminative domain adaptation","authors":"Eric Tzeng, Judy Hoffman, Kate Saenko, Trevor Darrell","year":"2017","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":33,"raw":"[Wang et al., 2019] Xiaojie Wang, Rui Zhang, Yu Sun, and Jianzhong Qi. Doubly robust joint learning for recom- mendation on data missing not at random. In ICML, pages 6638-6647, 2019.","title":"Doubly robust joint learning for recommendation on data missing not at random","authors":"Xiaojie Wang, Rui Zhang, Yu Sun, Jianzhong Qi","year":"2019","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":34,"raw":"[Wang et al., 2020] Zifeng Wang, Xi Chen, Rui Wen, Shao- Lun Huang, Ercan E. Kuruoglu, and Yefeng Zheng. In- formation theoretic counterfactual learning from missing- not-at-random feedback. In NeurIPS, 2020.","title":"Information theoretic counterfactual learning from missingnot-at-random feedback","authors":"Wang","year":"2020","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":35,"raw":"[Wang et al., 2022] Zimu Wang, Yue He, Jiashuo Liu, Wen- chao Zou, Philip S. Yu, and Peng Cui. Invariant prefer- ence learning for general debiasing in recommendation. In SIGKDD, pages 1969-1978, 2022.","title":"Invariant preference learning for general debiasing in recommendation","authors":"Wang","year":"2022","venue":"SIGKDD","doi":null,"url":null,"internal_url":null},{"position":36,"raw":"[Wu et al., 2019] Yifan Wu, Ezra Winston, Divyansh Kaushik, and Zachary C. Lipton. Domain adaptation with asymmetrically-relaxed distribution alignment. In ICML, pages 6872-6881, 2019.","title":"Domain adaptation with asymmetrically-relaxed distribution alignment","authors":"Yifan Wu, Ezra Winston, Divyansh Kaushik, Zachary C Lipton","year":"2019","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":37,"raw":"[Yan et al., 2017] Hongliang Yan, Yukang Ding, Peihua Li, Qilong Wang, Yong Xu, and Wangmeng Zuo. Mind the class weight bias: Weighted maximum mean discrepancy for unsupervised domain adaptation. In CVPR, pages 945- 954, 2017.","title":"Mind the class weight bias: Weighted maximum mean discrepancy for unsupervised domain adaptation","authors":"Hongliang Yan, Yukang Ding, Peihua Li, Qilong Wang, Yong Xu, Wangmeng Zuo","year":"2017","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":38,"raw":"[Zhang et al., 2019] Yuchen Zhang, Tianle Liu, Mingsheng Long, and Michael I. Jordan. Bridging theory and algo- rithm for domain adaptation. In ICML, pages 7404-7413, 2019.","title":"Bridging theory and algorithm for domain adaptation","authors":"Yuchen Zhang, Tianle Liu, Mingsheng Long, Michael I Jordan","year":"2019","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":39,"raw":"[Zhao et al., 2019] Han Zhao, Remi Tachet des Combes, Kun Zhang, and Geoffrey J. Gordon. On learning invari- ant representations for domain adaptation. In ICML, pages 7523-7532, 2019.","title":"On learning invariant representations for domain adaptation","authors":"Han Zhao, Remi Tachet Des Combes, Kun Zhang, Geoffrey J Gordon","year":"2019","venue":"ICML","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}