SIAM Student Chapter Seminar: Difference between revisions

From UW-Math Wiki
Jump to navigation Jump to search
Line 21: Line 21:
|Solly Parenti
|Solly Parenti
|What is ... a software engineering interview?
|What is ... a software engineering interview?
|-
|10/13
|Zoom and VV911
|Xiaopeng Li
|Convergence of the Momentum Method for Semi-Algebraic Functions with Locally Lipschitz Gradients
|-
|-
|10/20
|10/20
Line 47: Line 52:


'''September 29, Solly Parenti:''' I'll share my experiences going through a bunch of software engineering interviews, as well as how I learned how to program and my thoughts on industry jobs.
'''September 29, Solly Parenti:''' I'll share my experiences going through a bunch of software engineering interviews, as well as how I learned how to program and my thoughts on industry jobs.
'''October 13, Xiaopeng Li:'''  We propose a new length formula that governs the iterates of the momentum method when minimizing differentiable semi-algebraic functions with locally Lipschitz gradients. It enables us to establish local convergence, global convergence, and convergence to local minimizers without assuming global Lipschitz continuity of the gradient, coercivity, and a global growth condition, as is done in the literature. As a result, we provide the first convergence guarantee of the momentum method starting from arbitrary initial points when applied to principal component analysis, matrix sensing, and linear neural networks.
==Past Semesters==
==Past Semesters==
*[[SIAM Spring 2023]]
*[[SIAM Spring 2023]]

Revision as of 18:34, 8 October 2023


Fall2023

Date Location Speaker Title
9/29 Zoom and VV911 Solly Parenti What is ... a software engineering interview?
10/13 Zoom and VV911 Xiaopeng Li Convergence of the Momentum Method for Semi-Algebraic Functions with Locally Lipschitz Gradients
10/20 VV911 Yingxin Zhao
10/27 Zoom and VV911 Evan Sorensen
11/10 VV911 Jiayin Lu
11/17 VV911 Thomas Chandler


Abstracts

September 29, Solly Parenti: I'll share my experiences going through a bunch of software engineering interviews, as well as how I learned how to program and my thoughts on industry jobs.

October 13, Xiaopeng Li: We propose a new length formula that governs the iterates of the momentum method when minimizing differentiable semi-algebraic functions with locally Lipschitz gradients. It enables us to establish local convergence, global convergence, and convergence to local minimizers without assuming global Lipschitz continuity of the gradient, coercivity, and a global growth condition, as is done in the literature. As a result, we provide the first convergence guarantee of the momentum method starting from arbitrary initial points when applied to principal component analysis, matrix sensing, and linear neural networks.

Past Semesters