Joint Mathematics Meetings AMS Special Session
Current as of Saturday, January 18, 2025 03:30:04
- Program
- ·
- Deadlines
- ·
- Timetable
- ·
- Inquiries: meet@ams.org
2025 Joint Mathematics Meetings (JMM 2025)
- Seattle Convention Center and the Sheraton Grand Seattle, Seattle, WA
- January 8-11, 2025 (Wednesday - Saturday)
- Meeting #1203
Associate Secretary for the AMS Scientific Program:
Brian D. Boe, brian@math.uga.edu
AMS Special Session on Algebraic Methods in Machine Learning and Optimization
-
Friday January 10, 2025, 8:00 a.m.-12:00 p.m.
AMS Special Session on Algebraic Methods in Machine Learning and Optimization, I
This session will focus on applications of algebraic methods in understanding the mathematical theory of machine learning. It will cover topics including applied algebraic geometry, group symmetries, data and optimization invariances, and algebraic techniques to analyze the training and generalization of neural networks. The session will allow researchers to discuss and share their latest results on geometry of neural networks, equivariant architectures, symmetries, and non-convex optimization.
Skagit 2, Seattle Convention Center Arch at 800 Pike
Organizers:
Jiayi Li, University of California, Los Angeles jiayi.li@g.ucla.edu
Guido Francisco Montufar, MPI MiS
Yulia Alexandr, University of California, Los Angeles
Julia Lindberg, The University of Texas at Austin
-
8:00 a.m.
Introductory Remarks -
8:30 a.m.
On the Geometry and Optimization of Polynomial Convolutional Networks
Giovanni Luca Marchetti*, KTH Royal Institute of Technology
(1203-68-41877) -
9:00 a.m.
Generalization and Optimization in Symmetry-Preserving Machine Learning: Sample Complexity and Implicit Bias
Wei Zhu*, Georgia Institute of Technology
(1203-68-40220) -
9:30 a.m.
A transversality theorem for semi-algebraic sets
Dan Edidin*, Department of Mathematics, University of Missouri, Columbia, MO 65211
(1203-14-42654) -
10:00 a.m.
Decomposing Tensors via Rank-one Approximations
Emil Horobet, Sapientia Hungarian University of Transylvania
Alvaro Ribot*, Harvard University
Anna Seigal, Harvard University
Ettore Teixeira Turatti, UiT The Arctic University of Norway
(1203-15-43536) -
10:30 a.m.
The geometry of loss functions in machine learning
Yaim Cooper*, University of Notre Dame
(1203-51-44560) -
11:00 a.m.
Adversarially Robust Neural Network Decision Boundaries via Tropical Geometry
Jefferson Huang, Naval Postgraduate School, Monterey, CA
Keiji Miura, Kwansei Gakuin University, Sanda, Japan
Kurt Pasque, Naval Postgraduate School, Monterey, CA
Christopher Teska*, Naval Postgraduate School, Monterey, CA
Ruriko Yoshida, Naval Postgraduate School, Monterey, CA
(1203-14-43606) -
11:30 a.m.
On functional dimension and persistent pseudodimension
Julia Elisenda Grigsby*, Boston College
Kathryn Anne Lindsey, Boston College
(1203-68-43699)
-
8:00 a.m.
-
Friday January 10, 2025, 1:00 p.m.-6:00 p.m.
AMS Special Session on Algebraic Methods in Machine Learning and Optimization, II
This session will focus on applications of algebraic methods in understanding the mathematical theory of machine learning. It will cover topics including applied algebraic geometry, group symmetries, data and optimization invariances, and algebraic techniques to analyze the training and generalization of neural networks. The session will allow researchers to discuss and share their latest results on geometry of neural networks, equivariant architectures, symmetries, and non-convex optimization.
Skagit 2, Seattle Convention Center Arch at 800 Pike
Organizers:
Jiayi Li, University of California, Los Angeles jiayi.li@g.ucla.edu
Guido Francisco Montufar, MPI MiS
Yulia Alexandr, University of California, Los Angeles
Julia Lindberg, The University of Texas at Austin
-
1:00 p.m.
Approximation theory for graph neural networks via (sparse) graph limit
Thien Le*, Massachusetts Institute of Technology
(1203-68-43720) -
1:30 p.m.
Covering Number of Real Algebraic Varieties and Beyond: Improved Bounds and Applications
Joe Kileel, University of Texas at Austin
Yifan Zhang*, Oden Institute, University of Texas at Austin
(1203-14-43755) -
2:00 p.m.
Some complexity results for symmetric tensor decomposition over $\mathbb {C}$
Robert Shi*, The University of Texas at Austin
(1203-15-43523) -
2:30 p.m.
Break -
3:00 p.m.
Semialgebraic Methods in Convex Optimization
Benjamin D Grimmer, Johns Hopkins University
Kevin Shu*, Georgia Institute of Technology
Alex Wang, Perdue University
(1203-90-37965) -
3:30 p.m.
Relative Information and the Dual Numbers
Shaowei Lin*, Topos Institute
(1203-94-44056) -
4:00 p.m.
Discussion
-
1:00 p.m.