ICML全称为International Conference on Machine Learning（国际机器学习会议）。ICML 2021一共有5513篇有效论文投稿，最后1184篇被接收，被接收论文中有166篇论文被接收为Long Talk，1018篇论文被接收为Short Talk。

北京大学智能科学系林宙辰教授、王立威教授和王奕森助理教授有8篇论文被ICML 2021收录，包括2篇Long Talk。

被收录论文如下：

[1] Hongwei Wen, Jingyi Cui, Hanyuan Hang, Jiabin Liu,Yisen Wang, andZhouchen Lin, “Leveraged Weighted Loss for Partial Label Learning”，Long Talk

[2] Dinghuai Zhang, Kartik Ahuja, Yilun Xu,Yisen Wang, Aaron Courville, “Can Subnetwork Structure be the Key to Out-of-Distribution Generalization?”Long Talk

[3] Jingyi Cui, Hanyuan Hang,Yisen Wang, andZhouchen Lin, “GBHT: Gradient Boosting Histogram Transform for Density Estimation”

[4] Ruili Feng,Zhouchen Lin, Jiapeng Zhu, Deli Zhao, Jingren Zhou, and Zheng-Jun Zha, “Uncertainty Principles of Encoding GANs”

[5] Bohang Zhang, Tianle Cai, Zhou Lu, Di He,Liwei Wang, “Towards Certifying l-infty Robustness using Neural Networks with ℓ-infty-dist Neurons”

[6] Tianle Cai, Shengjie Luo, Keyulu Xu, Di He, Tie-Yan Liu,Liwei Wang,“GraphNorm: A Principled Approach to Accelerating Graph Neural Network Training”

[7] Jiachen Hu, Xiaoyu Chen, Chi Jin, Lihong Li,Liwei Wang, “Near-optimal Representation Learning for Linear Bandits and Linear RL”

[8] Tianhao Wu, Yunchang Yang, Simon S. Du,Liwei Wang, “On Reinforcement Learning with Adversarial Corruption and Its Application to Block MDP”