Machine Learning Group Seminar

Maintained by: Yi Xu

Location: B11 MacLean Hall (MLH)

Hours: 1:30pm - 3:30pm Friday

Week	Date	Topic	Presenter(s)

Week 1	Aug 24, 2018	None

Week 2	Aug 31, 2018	A Simple Nearly-Optimal Restart Scheme For Speeding-Up First Order Methods	Qihang Lin

Week 3	Sep 07, 2018	Chapters 2 and 3	Dixian Zhu and Xin Man

Week 4	Sep 14, 2018	Random Shuffling Beats SGD after Finite Epochs	Mingrui Liu

Week 5	Sep 21, 2018	Chapter 4	Hassan Rafique

Week 6	Sep 28, 2018	First-order Stochastic Algorithms for Escaping From Saddle Points in Almost Linear Time Fast Stochastic AUC Maximization with O(1/n) Convergence Rate (2:30pm-3:30pm)	Yi Xu and Mingrui Liu

Week 7	Oct 05, 2018	Chapters 5 and 6	Runchao Ma and Qi Qi

Week 8	Oct 12, 2018	Stochastic Convex Optimization: Faster Local Growth Implies Faster Global Convergence First-order Stochastic Algorithms for Escaping From Saddle Points in Almost Linear Time (2:30pm-3:30pm)	Yi Xu

Week 9	Oct 19, 2018	Gradient Descent Provably Optimizes Over-parameterized Neural Networks	Yi Xu

Week 10	Oct 26, 2018	Chapters 7 and 8	Zhishuai Guo and Mingrui Liu

Week 11	Nov 02, 2018	Variance-based Regularization with Convex Objectives	Yan Yan

Week 12	Nov 09, 2018	Improved Training of Wasserstein GANs	Zhuoning Yuan

Week 13	Nov 16, 2018	Adaptive Online Learning in Dynamic Environments	Xin Man

Week 14	Nov 23, 2018	Thanksgiving Break

Week 15	Nov 30, 2018	The Nonstochastic Multiarmed Bandit Problem KL-UCB-switch: optimal regret bounds for stochastic bandits from both a distribution-dependent and a distribution-free viewpoints	Dixian Zhu

Week 16	Dec 07, 2018		None

Week 17	Dec 14, 2018	Exam Week