Welcome to Kaiyi Ji's Homepage
 |
|
About Me
I am an assistant professor at the Department of Computer Science and Engineering of the University at Buffalo. I was a postdoctoral research fellow at the Electrical Engineering and Computer Science Department of the University of Michigan, Ann Arbor, in 2022, working with Prof. Lei Ying. I received my Ph.D. degree from the Electrical and Computer Engineering Department of The Ohio State University in December, 2021, advised by Prof. Yingbin Liang. I was a visiting student research collaborator at the department of Electrical Engineering, Princeton University working with Prof. H. Vincent Poor. Previously I obtained my B.S. degree from University of Science and Technology of China in 2016.
Research
I have been working at the intersection of optimization, machine learning and networked systems, on both the theory and application sides. My major research focuses include:
Bilevel optimization and its application in deep learning
Meta-learning and continual learning
Large-scale stochastic optimization
Federated learning and communication networks
To Prospective Students
I am looking for highly motivated students with strong mathematical backgrounds and/or programming skills in machine learning and optimization to work with me.
Moreover, intern and visiting students are highly welcome!
One PhD position is avaliable immediately. Please fill the following form if you are interested. I will contact you if there is a good match!
News!
09/2023 Five papers accepted in NeurIPS 2023 with one spotlight presentation! The topics span over Hessian-free bilevel optimization, federated learning, continual learning and multi-objective learning. Big congratulations to my students Yifan, Peiyao and Hao, and many thanks to my collaborators!
07/2023 I received a grant as site PI @ UB in collaboration with Dr. Shiqian Ma (Rice University) on Distributed Optimization in Multi-Agent Systems. Many thanks to NSF!
07/2023 I received a grant as Lead PI @ UB in collaboration with Dr. Shiqian Ma (Rice University) on Large-Scale Bilevel Optimization. Many thanks to NSF!
07/2023 Our work on fair resource allocation over communication network has been accepted by ACM MobiHoc 2023! (acceptance rate: 21.9%)
06/2023 Gave an invited talk on bilevel optimization and continual learning at SIAM Conference on Optimization.
06/2023 Two new manuscripts on Simple federated bilevel learning and Direction-oriented multi-objective/task learning are available online. Codes can be found in GitHub.
04/2023 Our two submissions on federated bilevel optimization: Achieving Linear Speedup and Aggregated Iterative Differentiation have been accepted at ICML 2023 (acceptance rate: 27.9%). Congrats to my students Peiyao and Dewei! This is also Peiyao's first work. Keep going!
02/2023 Selected as Top Reviewer in AISTATS 2023.
01/2023 A manuscript on fair resource allocation over communication network is available online: Network Utility Maximization with Bilevel Learning. We propose a new distributed data-driven bilevel optimization approach for network utility maximization with unknown user utilities!
11/2022 Our NeurIPS 2022 paper on Unified Analysis for Bilevel Optimization received Spotlight (5% acceptance rate).
10/2022 Will serve as TPC member of ACM MobiHoc 2023. Please consider submitting your best work to the conference!
10/2022 Gave an invited talk in the session “Bilevel Stochastic Methods for Optimization and Learning” at the 2022 INFORMS Annual Meeting, Indianapolis, IN.
09/2022 Two papers on Hessian-Free Bilevel Optimization and Unified Analysis for Bilevel Optimization have been accepted by NeurIPS 2022.
09/2022 Our paper, “Lower Bounds and Accelerated Algorithms for Bilevel Optimization”, has been accepted by JMLR 2022 with minor revision.
New manuscript out: Will Bilevel Optimizers Benefit from Loops. We develop a unified convergence theory for AID- and ITD-based bilevel optimization, which covers all implementation choices of loops. Check it out!
New manuscript out: Efficiently Escaping Saddle Points in Bilevel Optimization. We develop first-order methods that can efficiently escape saddle points in minimax and bilevel optimization. Joint work with Minhui Huang, Shiqian Ma, Lifeng Lai from UC Davis.
New manuscript out: A Constrained Optimization Approach to Bilevel Optimization with Multiple Inner Minima. We develop a primal-dual Hessian-free bilevel method with the first-known convergence rate analysis in problems with multiple inner minima!
Prof. Yingbin Liang (from OSU) and I are organizing an invited session on “Bilevel Machine Learning” for the 2022 Conference on Information Sciences and Systems (CISS) virtually from March 9 to 11, 2022. see <website> for more details.
12/2021 Invited talk at the Next Generation Transportation Systems (NGTS) Seminar, University of Michigan.
I am honored to receive an outstanding reviewer award from NeurIPS 2021 (top 8% of reviewers)!
Our recent paper on momentum-based bilevel optimization has been accepted by NeurIPS 2021 with a spotlight presentation (3% acceptance rate)! Check our paper and code. More results will be added.
I am rewarded as expert reviewer and top 10% reviewer for ICML 2021 <Link>! Really honered!
Our paper on bilevel optimization has been accepted by ICML 2021. We provide comprehensive nonasympototic analysis and stochastic solver stocBiO for this area. Check our paper and code!
Our paper on theoretical analysis of MAML has been accepted by JMLR 2021.
Our paper on generalization error of GANs has been accepted by Trans. on Information Theory (TIT) 2021.
I am really honored to receive the 2020/21 Presidential Fellowship at OSU, the highest award for graduate students!
I was awarded as a top 10% high-scoring reviewer for NeurIPS 2020. Thanks!