Yi Xu, PhD

Professor
Dalian University of Technology

Address: 2 Ling Gong Road, Gan Jing Zi District, Dalian, Liaoning 116024, China
Email: yxu AT dlut DOT edu DOT cn

I am a professor at Dalian University of Technology. I was a machine learning researcher at Alibaba DAMO Academy. I received my Ph.D. degree in Computer Science from Department of Computer Science, The University of Iowa in 2019. I was very fortunate to have Prof. Tianbao Yang as my advisor and Prof. Qihang Lin as my co-advisor. My Ph.D. dissertation received the Spring 2019 Ballard and Seashore Dissertation Fellowships at the University of Iowa. Prior to that, I received my master's degree from South Dakota State University and my bachelor's degree from Zhejiang University. Here is my Google Scholar Citations.

Research Interests

News

(Mar 2022) Two papers were accepted to CVPR 2022.
(Feb 2022) One paper was accepted to ICLR 2022.
(Sep 2021) One paper was accepted to NeurIPS 2021.
(May 2021) Two papers were accepted to ICML 2021.

Preprints

(New!) Improved Fine-tuning by Leveraging Pre-training Data: Theory and Practice
Ziquan Liu*, Yi Xu*, Yuanhong Xu, Qi Qian, Hao Li, Antoni Chan, Rong Jin. (*equal contribution).
arXiv Preprint: 2111.12292
(New!) Self-Supervised Pre-Training for Transformer-Based Person Re-Identification
Hao Luo, Pichao Wang, Yi Xu, Feng Ding, Yanxin Zhou, Fan Wang, Hao Li, Rong Jin.
arXiv Preprint: 2111.12084
(New!) On Stochastic Moving-Average Estimators for Non-Convex Optimization
Zhishuai Guo, Yi Xu, Wotao Yin, Rong Jin, Tianbao Yang.
arXiv Preprint: 2104.14840
(New!) A Convergence Theory Towards Practical Over-parameterized Deep Neural Networks
Asaf Noy*, Yi Xu*, Yonathan Aflalo*, Lihi Zelnik-Manor*, Rong Jin*. (*equal contribution)
arXiv Preprint: 2101.04243
Attentional Biased Stochastic Gradient for Imbalanced Classification
Qi Qi, Yi Xu, Rong Jin, Wotao Yin, Tianbao Yang.
arXiv Preprint: 2012.06951
WeMix: How to Better Utilize Data Augmentation
Yi Xu, Asaf Noy, Ming Lin, Qi Qian, Hao Li, Rong Jin.
arXiv Preprint:2010.01267
Towards Understanding Label Smoothing
Yi Xu, Yuanhong Xu, Qi Qian, Hao Li, Rong Jin.
arXiv Preprint:2006.11653
Stochastic Primal-Dual Algorithms with Faster Convergence than O(1/√T) for Problems without Bilinear Structure
Yan Yan, Yi Xu, Qihang Lin, Lijun Zhang, Tianbao Yang.
arXiv Preprint:1904.10112
(TR) Frank-Wolfe Method is Automatically Adaptive to Error Bound Condition
Yi Xu, Tianbao Yang.
arXiv Preprint:1810.04765

Publications

(New!) Learning from Untrimmed Videos: Self-Supervised Video Representation Learning with Hierarchical Consistency
Zhiwu Qing, Shiwei Zhang, Ziyuan Huang, Yi Xu, Xiang Wang, Mingqian Tang, Changxin Gao, Rong Jin, Nong Sang.
Accepted to CVPR 2022.
(New!) CHEX: CHannel EXploration for CNN Model Compression
Zejiang Hou, Minghai Qin, Fei Sun, Xiaolong Ma, Kun Yuan, Yi Xu, Yen-Kuang Chen, Rong Jin, Yuan Xie, Sun-Yuan Kung.
Accepted to CVPR 2022.
(New!) Effective Model Sparsification by Scheduled Grow-and-Prune Methods
Xiaolong Ma, Minghai Qin, Fei Sun, Zejiang Hou, Kun Yuan, Yi Xu, Yanzhi Wang, Yen-Kuang Chen, Rong Jin, Yuan Xie.
Accepted to the Tenth International Conference on Learning Representations, 2022. (ICLR 2022)
(New!) An Online Method for A Class of Distributionally Robust Optimization with Non-convex Objectives
Qi Qi, Zhishuai Guo, Yi Xu, Rong Jin, Tianbao Yang.
Advances In Neural Information Processing Systems 34, 2021. (NeurIPS 2021)
Dash: Semi-Supervised Learning with Dynamic Thresholding
Yi Xu, Lei Shang, Jinxing Ye, Qi Qian, Yu-Feng Li, Baigui Sun, Hao Li, Rong Jin.
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:11525-11536, 2021. (ICML 2021) (Long Talk, acceptance rate: 3%)
[Python Code]
Federated Deep AUC Maximization for Heterogeneous Data with a Constant Communication Complexity
Zhuoning Yuan*, Zhishuai Guo*, Yi Xu, Yiming Ying, Tianbao Yang. (*equal contribution)
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:12219-12229, 2021. (ICML 2021)
Optimal Epoch Stochastic Gradient Descent Ascent Methods for Min-Max Optimization
Yan Yan, Yi Xu, Qihang Lin, Wei Liu, Tianbao Yang.
Advances In Neural Information Processing Systems 33, 5789--5800, 2020. (NeurIPS 2020)
Stochastic Optimization for Non-convex Inf-Projection Problems
Yan Yan, Yi Xu, Lijun Zhang, Xiaoyu Wang, Tianbao Yang.
Proceedings of the 37th International Conference on Machine Learning, PMLR 119: 10660-10669, 2020. (ICML 2020)
Non-asymptotic Analysis of Stochastic Methods for Non-Smooth Non-Convex Regularized Problems
Yi Xu, Rong Jin, Tianbao Yang.
Advances In Neural Information Processing Systems 32, 2630--2640, 2019. (NeurIPS 2019)
Learning with Non-Convex Truncated Losses by SGD
Yi Xu, Shenghuo Zhu, Sen Yang, Chi Zhang, Rong Jin, Tianbao Yang.
Proceedings of Conference on Uncertainty in Artificial Intelligence, 2019. (UAI 2019)
On the Convergence of (Stochastic) Gradient Descent with Extrapolation for Non-Convex Optimization
Yi Xu, Zhuoning Yuan, Sen Yang, Rong Jin, Tianbao Yang.
Proceedings of the 28th International Joint Conference on Artificial Intelligence, 4003-4009, 2019. (IJCAI 2019)
Stochastic Optimization for DC Functions and Non-smooth Non-convex Regularizers with Non-asymptotic Convergence
Yi Xu, Qi Qi, Qihang Lin, Rong Jin, Tianbao Yang.
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:6942-6951, 2019. (ICML 2019)
Katalyst: Boosting Convex Katayusha for Non-Convex Problems with a Large Condition Number
Zaiyi Chen, Yi Xu, Haoyuan Hu, Tianbao Yang.
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:1102-1111, 2019. (ICML 2019)
Accelerate Stochastic Subgradient Method by Leveraging Local Growth Condition
Yi Xu, Qihang Lin, Tianbao Yang.
Analysis and Applications Vol. 17, No. 05, pp. 773-818 (2019) [DOI]
First-order Stochastic Algorithms for Escaping From Saddle Points in Almost Linear Time [Neon and Neon+]
Yi Xu, Rong Jin, Tianbao Yang.
Advances In Neural Information Processing Systems 31, 5535-5545, 2018. (NeurIPS 2018)
SADAGRAD: Strongly Adaptive Stochastic Gradient Methods
Zaiyi Chen*, Yi Xu*, Enhong Chen, Tianbao Yang. (*equal contribution)
Proceedings of the 35th International Conference on Machine Learning, PMLR 80: 913-921, 2018. (ICML 2018)
[Supplement]
Adaptive SVRG Methods under Error Bound Conditions with Unknown Growth Parameter
Yi Xu, Qihang Lin, Tianbao Yang.
Advances In Neural Information Processing Systems 30, 3279-3289, 2017. (NeurIPS 2017)
[Supplement] [Poster] [Bibtex]
ADMM without a Fixed Penalty Parameter: Faster Convergence with New Adaptive Penalization
Yi Xu, Mingrui Liu, Qihang Lin, Tianbao Yang.
Advances In Neural Information Processing Systems 30, 1267-1277, 2017. (NeurIPS 2017)
[Supplement] [Poster] [Bibtex]
Stochastic Convex Optimization: Faster Local Growth Implies Faster Global Convergence
Yi Xu, Qihang Lin, Tianbao Yang.
Proceedings of the 34th International Conference on Machine Learning, PMLR 70: 3821-3830, 2017. (ICML 2017)
(short version) In NeurIPS workshop on Optimization for Machine Learning, 2016
[Supplement] [arXiv Version][Matlab Code] [Bibtex]
Efficient Non-oblivious Randomized Reduction for Risk Minimization with Improved Excess Risk Guarantee
Yi Xu, Haiqing Yang, Lijun Zhang, Tianbao Yang.
Proceedings of the 31st AAAI Conference on Artificial Intelligence, 2796--2802, 2017. (AAAI 2017)
[with Supplement] [Bibtex]
Homotopy Smoothing for Non-Smooth Problems with Lower Complexity than O(1/ε)
Yi Xu*, Yan Yan*, Qihang Lin, Tianbao Yang. (*equal contribution)
Advances In Neural Information Processing Systems 29, 1208--1216, 2016. (NeurIPS 2016)
[Supplement] [arXiv Version] [Bibtex]

Services

PC Memeber
The 34th AAAI Conference on Artificial Intelligence (AAAI 2020)
The 28th International Joint Conference on Artificial Intelligence (IJCAI 2019)
The 33rd AAAI Conference on Artificial Intelligence (AAAI 2019)
The 27th International Joint Conference on Artificial Intelligence and the 23rd European Conference on Artificial Intelligence (IJCAI-ECAI 2018)
Reviewer
Advances in Neural Information Processing System (NeurIPS 2019)
The 31st AAAI Conference on Artificial Intelligence (AAAI 2017)
Advances in Neural Information Processing System (NIPS 2016)
The 25th International Joint Conference on Artificial Intelligence (IJCAI 2016)
ACM Transactions on Knowledge Discovery from Data
Neurocomputing
Frontiers of Computer Science
IEEE Transactions on Signal Processing
IEEE Transactions on Big Data
IEEE Access
Session Chair
SIAM Conference on Optimization (OP17), Vancouver, BC, Canada, 2017

Miscellaneous

Last update: Mar 3, 2022