{"type":"ScholarlyArticle","title":"Curriculum Multi-Level Learning for Imbalanced Live-Stream Recommendation","authors":[{"name":"Shuodian Yu","slug":"shuodian-yu","given":"Shuodian","family":"Yu","affiliation":"MoE Key Lab of Artificial Intelligence , Department of Computer Science and Engineering , Shanghai Jiao Tong University , Shanghai , China"},{"name":"Junqi Jin","slug":"junqi-jin","given":"Junqi","family":"Jin","affiliation":"Alibaba Group , Beijing , China"},{"name":"Li Ma","slug":"li-ma","given":"Li","family":"Ma","affiliation":"MoE Key Lab of Artificial Intelligence , Department of Computer Science and Engineering , Shanghai Jiao Tong University , Shanghai , China"},{"name":"Xiaofeng Gao","slug":"xiaofeng-gao","given":"Xiaofeng","family":"Gao","affiliation":"MoE Key Lab of Artificial Intelligence , Department of Computer Science and Engineering , Shanghai Jiao Tong University , Shanghai , China"},{"name":"Xiaopeng Wu","slug":"xiaopeng-wu","given":"Xiaopeng","family":"Wu","affiliation":"Alibaba Group , Beijing , China"},{"name":"Haiyang Xu","slug":"haiyang-xu","given":"Haiyang","family":"Xu","affiliation":"Alibaba Group , Beijing , China"},{"name":"Jian Xu","slug":"jian-xu","given":"Jian","family":"Xu","affiliation":"Alibaba Group , Beijing , China"}],"abstract":"In large-scale e-commerce live-stream recommendation, streamers are classified into different levels based on their popularity and other metrics for marketing. Several top streamers at the head level occupy a considerable amount of exposure, resulting in an unbalanced data distribution. A unified model for all levels without consideration of imbalance issue can be biased towards head streamers and neglect the conflicts between levels. The lack of inter-level streamer correlations and intra-level streamer characteristics modeling imposes obstacles to estimating the user behaviors. To tackle these challenges, we propose a curriculum multi-level learning framework for imbalanced recommendation. We separate model parameters into shared and level-specific ones to explore the generality among all levels and discrepancy for each level respectively. The level-aware gradient descent and a curriculum sampling scheduler are designed to capture the de-biased commonalities from all levels as the shared parameters. During the specific parameters training, the hardness-aware learning rate and an adaptor are proposed to dynamically balance the training process. Finally, shared and specific parameters are combined to be the final model weights and learned in a cooperative training framework. Extensive experiments on a live-stream production dataset demonstrate the superiority of the proposed framework.","doi":"10.24963/ijcai.2023/267","url":"https://ijcai.org/proceedings/2023/267","pdf":"https://ijcai.org/proceedings/2023/0267.pdf","markdown":"https://ijcai.org/proceedings/2023/267.md","bibtex":"https://ijcai.org/proceedings/2023/bibtex/267","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":267,"pages":{"first":2406,"last":2414},"keywords":["Data Mining: DM: Recommender systems","Machine Learning: ML: Multi-task and transfer learning"],"videos":[],"references":[{"position":1,"raw":"[Allison, 1978] Paul D Allison. Measures of inequality. American sociological review, pages 865-880, 1978.","title":"Measures of inequality","authors":"Paul D Allison","year":"1978","venue":"American sociological review","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Alshammari et al., 2022] Shaden Alshammari, Yu-Xiong Wang, Deva Ramanan, and Shu Kong. Long-tailed recog- nition via weight balancing. In CVPR, pages 6897-6907, 2022.","title":"Long-tailed recognition via weight balancing","authors":"Shaden Alshammari, Yu-Xiong Wang, Deva Ramanan, Shu Kong","year":"2022","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Caruana, 1997] Rich Caruana. Multitask learning. Machine learning, 28(1):41-75, 1997.","title":"Multitask learning","authors":"Caruana","year":"1997","venue":"Machine learning","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Chen et al., 2018] Zhao Chen, Vijay Badrinarayanan, Chen-Yu Lee, and Andrew Rabinovich. Gradnorm: Gradient normalization for adaptive loss balancing in deep multitask networks. In ICML, pages 794-803.","title":"Gradnorm: Gradient normalization for adaptive loss balancing in deep multitask networks","authors":"Vijay Zhao Chen, Chen-Yu Badrinarayanan, Andrew Lee, Rabinovich","year":null,"venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"PMLR, 2018.","title":null,"authors":"Pmlr","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Chen et al., 2021] Yudong Chen, Xin Wang, Miao Fan, Jizhou Huang, Shengwen Yang, and Wenwu Zhu. Cur- riculum meta-learning for next poi recommendation. In SIGKDD, pages 2692-2702, 2021.","title":"Curriculum meta-learning for next poi recommendation","authors":"Yudong Chen, Xin Wang, Miao Fan, Jizhou Huang, Shengwen Yang, Wenwu Zhu","year":"2021","venue":"SIGKDD","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Cui et al., 2019] Yin Cui, Menglin Jia, Tsung-Yi Lin, Yang Song, and Serge Belongie. Class-balanced loss based on effective number of samples. In CVPR, pages 9268-9277, 2019.","title":"Class-balanced loss based on effective number of samples","authors":"Yin Cui, Menglin Jia, Tsung-Yi Lin, Yang Song, Serge Belongie","year":"2019","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Dredze et al., 2010] Mark Dredze, Alex Kulesza, and Koby Crammer. Multi-domain learning by confidence-weighted parameter combination. Machine Learning, 79(1):123- 149, 2010.","title":"Multi-domain learning by confidence-weighted parameter combination","authors":"Mark Dredze, Alex Kulesza, Koby Crammer","year":"2010","venue":"Machine Learning","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Finn et al., 2017] Chelsea Finn, Pieter Abbeel, and Sergey Levine. Model-agnostic meta-learning for fast adaptation of deep networks. In International conference on machine learning, pages 1126-1135. PMLR, 2017.","title":"Model-agnostic meta-learning for fast adaptation of deep networks","authors":"Finn","year":"2017","venue":"International conference on machine learning","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Jamal and Qi, 2019] Muhammad Abdullah Jamal and Guo- Jun Qi. Task agnostic meta-learning for few-shot learning. In ICCV, pages 11719-11727, 2019.","title":"Muhammad Abdullah Jamal and Guo-Jun Qi. Task agnostic meta-learning for few-shot learning","authors":"Qi Jamal","year":"2019","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Kang et al., 2020] Bingyi Kang, Saining Xie, Marcus Rohrbach, Zhicheng Yan, Albert Gordo, Jiashi Feng, and Yannis Kalantidis. Decoupling representation and clas- sifier for long-tailed recognition. In ICLR, pages 1-16, 2020.","title":"Decoupling representation and classifier for long-tailed recognition","authors":"Kang","year":"2020","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Kingma and Ba, 2014] Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.","title":"Adam: A method for stochastic optimization","authors":"P Diederik, Jimmy Kingma, Ba","year":"2014","venue":null,"doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Li et al., 2020] Pengcheng Li, Runze Li, Qing Da, An- Xiang Zeng, and Lijun Zhang. Improving multi-scenario learning to rank in e-commerce by exploiting task relation- ships in the label space. In CIKM, pages 2605-2612, 2020.","title":"Improving multi-scenario learning to rank in e-commerce by exploiting task relationships in the label space","authors":"Pengcheng Li, Runze Li, Qing Da, An-Xiang Zeng, Lijun Zhang","year":"2020","venue":"CIKM","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Luo et al., 2022] Linhao Luo, Yumeng Li, Buyu Gao, Shuai Tang, Sinan Wang, Jiancheng Li, Tanchao Zhu, Jiancai Liu, Zhao Li, Binqiang Zhao, et al. Mamdr: A model agnostic learning method for multi-domain recommenda- tion. arXiv preprint arXiv:2202.12524, 2022.","title":"Mamdr: A model agnostic learning method for multi-domain recommendation","authors":"Luo","year":"2022","venue":null,"doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Ma et al., 2018a] Jiaqi Ma, Zhe Zhao, Xinyang Yi, Jilin Chen, Lichan Hong, and Ed H Chi. Modeling task rela- tionships in multi-task learning with multi-gate mixture- of-experts. In SIGKDD, pages 1930-1939, 2018.","title":"Modeling task relationships in multi-task learning with multi-gate mixtureof-experts","authors":"Ma","year":"2018","venue":"SIGKDD","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Ma et al., 2018b] Xiao Ma, Liqin Zhao, Guan Huang, Zhi Wang, Zelin Hu, Xiaoqiang Zhu, and Kun Gai. Entire space multi-task model: An effective approach for esti- mating post-click conversion rate. In SIGIR, pages 1137- 1140, 2018.","title":"Entire space multi-task model: An effective approach for estimating post-click conversion rate","authors":"Ma","year":"2018","venue":"SIGIR","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Misra et al., 2016] Ishan Misra, Abhinav Shrivastava, Abhi- nav Gupta, and Martial Hebert. Cross-stitch networks for multi-task learning. In ICCV, pages 3994-4003, 2016.","title":"Cross-stitch networks for multi-task learning","authors":"Ishan Misra, Abhinav Shrivastava, Abhinav Gupta, Martial Hebert","year":"2016","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Nair and Hinton, 2010] Vinod Nair and Geoffrey E Hinton. Rectified linear units improve restricted boltzmann ma- chines. In ICML, 2010.","title":"Rectified linear units improve restricted boltzmann machines","authors":"Vinod Nair, Geoffrey E Hinton","year":"2010","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Nichol et al., 2018] Alex Nichol, Joshua Achiam, and John Schulman. On first-order meta-learning algorithms. arXiv preprint arXiv:1803.02999, 2018.","title":"On first-order meta-learning algorithms","authors":"Alex Nichol, Joshua Achiam, John Schulman","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Sheng et al., 2021] Xiang-Rong Sheng, Liqin Zhao, Guorui Zhou, Xinyao Ding, Binding Dai, Qiang Luo, Siran Yang, Jingshan Lv, Chi Zhang, Hongbo Deng, et al. One model to serve all: Star topology adaptive recommender for multi-domain ctr prediction. In CIKM, pages 4104-4113, 2021.","title":"One model to serve all: Star topology adaptive recommender for multi-domain ctr prediction","authors":"Xiang-Rong Sheng, Liqin Zhao, Guorui Zhou, Xinyao Ding, Binding Dai, Qiang Luo, Siran Yang, Jingshan Lv, Chi Zhang, Hongbo Deng","year":"2021","venue":"CIKM","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Tang et al., 2020] Hongyan Tang, Junning Liu, Ming Zhao, and Xudong Gong. Progressive layered extraction (ple): A novel multi-task learning (mtl) model for personalized recommendations. In RecSys, pages 269-278, 2020.","title":"Progressive layered extraction (ple): A novel multi-task learning (mtl) model for personalized recommendations","authors":"Hongyan Tang, Junning Liu, Ming Zhao, Xudong Gong","year":"2020","venue":"RecSys","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Thrun and Pratt, 2012] Sebastian Thrun and Lorien Pratt. Learning to learn. Springer Science & Business Media, 2012.","title":"Sebastian Thrun and Lorien Pratt","authors":"Pratt Thrun","year":"2012","venue":null,"doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Wang et al., 2019] Yiru Wang, Weihao Gan, Jie Yang, Wei Wu, and Junjie Yan. Dynamic curriculum learning for im- balanced data classification. In ICCV, pages 5017-5026, 2019.","title":"Dynamic curriculum learning for imbalanced data classification","authors":"Wang","year":"2019","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Xi et al., 2021] Dongbo Xi, Zhen Chen, Peng Yan, Yinger Zhang, Yongchun Zhu, Fuzhen Zhuang, and Yu Chen. Modeling the sequential dependence among audience multi-step conversions with multi-task learning in targeted display advertising. In SIGKDD, 2021.","title":"Modeling the sequential dependence among audience multi-step conversions with multi-task learning in targeted display advertising","authors":"Dongbo Xi, Zhen Chen, Peng Yan, Yinger Zhang, Yongchun Zhu, Fuzhen Zhuang, Yu Chen","year":"2021","venue":"SIGKDD","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Yu et al., 2020] Tianhe Yu, Saurabh Kumar, Abhishek Gupta, Sergey Levine, Karol Hausman, and Chelsea Finn. Gradient surgery for multi-task learning. NeurIPS, 33:5824-5836, 2020.","title":"Gradient surgery for multi-task learning","authors":"Yu","year":"2020","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Zhang and Yang, 2021] Yu Zhang and Qiang Yang. A sur- vey on multi-task learning. TKDE, 2021.","title":"A survey on multi-task learning","authors":"Yu Zhang, Qiang Yang","year":"2021","venue":"TKDE","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Zhang et al., 2021a] Yifan Zhang, Bingyi Kang, Bryan Hooi, Shuicheng Yan, and Jiashi Feng. Deep long-tailed learning: A survey. arXiv preprint arXiv:2110.04596, 2021.","title":"Deep long-tailed learning: A survey","authors":"Zhang","year":"2021","venue":null,"doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Zhang et al., 2021b] Yin Zhang, Derek Zhiyuan Cheng, Tiansheng Yao, Xinyang Yi, Lichan Hong, and Ed H Chi. A model of two tales: Dual transfer learning framework for improved long-tail item recommendation. In WWW, pages 2220-2231, 2021.","title":"A model of two tales: Dual transfer learning framework for improved long-tail item recommendation","authors":"Zhang","year":"2021","venue":"WWW","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Zhang et al., 2022a] Qianqian Zhang, Xinru Liao, Quan Liu, Jian Xu, and Bo Zheng. Leaving no one behind: a multi-scenario multi-task meta learning approach for ad- vertiser modeling. In WSDM, pages 1368-1376, 2022.","title":"Leaving no one behind: a multi-scenario multi-task meta learning approach for advertiser modeling","authors":"Zhang","year":"2022","venue":"WSDM","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Zhang et al., 2022b] Yuanxing Zhang, Langshi Chen, Siran Yang, Man Yuan, Huimin Yi, Jie Zhang, Jiamang Wang, Jianbo Dong, Yunlong Xu, Yue Song, et al. Picasso: Un- leashing the potential of gpu-centric training for wide-and- deep recommender systems. In ICDE, pages 3453-3466. IEEE, 2022.","title":"Picasso: Unleashing the potential of gpu-centric training for wide-anddeep recommender systems","authors":"Zhang","year":"2022","venue":"ICDE","doi":null,"url":null,"internal_url":null},{"position":31,"raw":"[Zhou et al., 2020] Boyan Zhou, Quan Cui, Xiu-Shen Wei, and Zhao-Min Chen. Bbn: Bilateral-branch network with cumulative learning for long-tailed visual recognition. In CVPR, pages 9719-9728, 2020.","title":"Bbn: Bilateral-branch network with cumulative learning for long-tailed visual recognition","authors":"Boyan Zhou, Quan Cui, Xiu-Shen Wei, Zhao-Min Chen","year":"2020","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":32,"raw":"[Zou et al., 2022] Xinyu Zou, Zhi Hu, Yiming Zhao, Xuchu Ding, Zhongyi Liu, Chenliang Li, and Aixin Sun. Au- tomatic expert selection for multi-scenario and multi-task search. In SIGIR, page 1535-1544, 2022.","title":"Automatic expert selection for multi-scenario and multi-task search","authors":"Xinyu Zou, Zhi Hu, Yiming Zhao, Xuchu Ding, Zhongyi Liu, Chenliang Li, Aixin Sun","year":"2022","venue":"SIGIR","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}