Machine Learning Group Seminar (2021 Spring)

Time: 2:30PM - 4:30 PM CST, Friday, Online: Zoom link

Hosted by: Tianbao Yang, Qihang Lin, Peng Jiang

Topic: The general topics of interest is machine learning and optimization. In particular, we are interested in stochastic optimization, non-convex optimization, constrained optimization, efficient methods for large-scale training of deep neural networks, etc.

Maintained by: Qi Qi

Week	Date	Topic	Presenter	Recorded Videos
1	Jan 29, 2021	Canceled for ICML2021
2	Feb 5, 2021	Canceled for ICML2021
3	Feb 12, 2021	Fine-Grained Analysis of Stability and Generalization for Stochastic Gradient Descent	Quanqi Hu	Recorded Video
4	Feb 19, 2021	A PRIMAL-DUAL ALGORITHM FOR GENERAL CONVEX-CONCAVE SADDLE POINT PROBLEMS	Yao Yao	Recorded Video
5	Feb 26, 2021	Lower Bounds and Optimal Algorithms for Personalized Federated Learning	Zhishuai Guo	Recorded Video
6	Mar 5, 2021	Meta Pseudo Labels	Dixian Zhu	Recorded Video
7	Mar 12, 2021	A Single-Timescale Stochastic Bilevel Optimization Method	Qi Qi	Recorded Video
8	Mar 19, 2021	A Theoretical Analysis of Contrastive Unsupervised Representation Learning	Xin Man
9	Mar 26, 2021	On First-Order Meta-Learning Algorithms	Zhuoning Yuan
10	Apr 2, 2021	Improving the Sample and Communication Complexity for Decentralized Non-Convex Optimization: Joint Gradient Estimation and Tracking	Yankun Huang
11	Apr 9, 2021	A Two-Timescale Stochastic Algorithm Framework for Bilevel Optimization: Complexity Analysis and Application to Actor-Critic	Quanqi Hu	Recorded Video
12	Apr 16, 2021	Variance Reduction via Primal-Dual Accelerated Dual Averaging for Nonsmooth Convex Finite-Sums	Yao Yao
13	Apr 23, 2021	Stochastic optimization with decision-dependent distributions	Zhishuai Guo
14	Apr 30, 2021	Robust Optimization for Fairness with Noisy Protected Groups	Qi Qi
15	May 7, 2021	Wide Neural Networks of Any Depth Evolve as Linear Models Under Gradient Descent	Dixian Zhu
15	May 14, 2021		Final Exam Week