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 Topological, Algebraic, and Geometric Methods for Safe, Robust, and Explainable Machine Learning

• Wednesday January 8, 2025, 8:00 a.m.-12:00 p.m.
  AMS Special Session on Topological, Algebraic, and Geometric Methods for Safe, Robust, and Explainable Machine Learning, I
  This special session showcases research that applies ideas from topology, algebra, and geometry to the goal of increasing the safety, robustness, or explainability of modern machine learning. We will feature research that (i) proposes novel approaches to machine learning by drawing on tools and ideas from topology, algebra, and geometry or (ii) uses mathematics to illuminate how and why existing state-of-the-art models work as well as they do in some situations but fail in others.
  613, Seattle Convention Center Arch at 705 Pike
  Organizers:
  Henry Kvinge, Pacific Northwest National Laboratory henry.kvinge@pnnl.gov
  Tegan Emerson, Pacific Northwest National Laboratory
  Tim Doster, Pacific Northwest National Lab
  Scott Mahan, Pacific Northwest National Laboratory
  Sarah McGuire, Michigan State University
  

  • 8:00 a.m.
    A Neural Net Model for Distillation with Weights Explained
    Amire Bendjeddou, EPFL
    Johanni Brea, EPFL
    Berfin Simsek*, Flatiron Institute & NYU
    (1203-82-45439)
  • 8:30 a.m.
    Analysis of internal activations to indicate undesirable behaviors in large language models
    Jonathan H Tu*, Pacific Northwest National Laboratory
    (1203-68-41288)
  • 9:00 a.m.
    Interpreting Neural Networks Trained on Combinatorial Data
    Herman Chau*, University of Washington
    (1203-68-41832)
  • 9:30 a.m.
    Geometric guarantees for explainable data science
    Vitaliy A Kurlin*, University of Liverpool (UK)
    (1203-51-36844)
  • 10:00 a.m.
    Break
  • 10:30 a.m.
    Diss-lECT: Dissecting Data with local Euler Characteristic Transforms
    Bastian Rieck*, University of Fribourg
    (1203-55-40295)
  • 11:00 a.m.
    Modeling Many-to-Many Maps
    Elizabeth Diane Coda*, Pacific Northwest National Laboratory (PNNL)
    (1203-68-42370)
  • 11:30 a.m.
    $O(k)$-Equivariant Dimensionality Reduction on Stiefel Manifolds
    Andrew Lee, St. Thomas Aquinas College
    Harlin Lee*, University of North Carolina at Chapel Hill
    Jose Perea, Northeastern University
    Nikolas Schonsheck, Rockefeller University
    Madeleine Weinstein, University of Puget Sound
    (1203-55-39682)
• Wednesday January 8, 2025, 1:00 p.m.-5:00 p.m.
  AMS Special Session on Topological, Algebraic, and Geometric Methods for Safe, Robust, and Explainable Machine Learning, II
  This special session showcases research that applies ideas from topology, algebra, and geometry to the goal of increasing the safety, robustness, or explainability of modern machine learning. We will feature research that (i) proposes novel approaches to machine learning by drawing on tools and ideas from topology, algebra, and geometry or (ii) uses mathematics to illuminate how and why existing state-of-the-art models work as well as they do in some situations but fail in others.
  613, Seattle Convention Center Arch at 705 Pike
  Organizers:
  Henry Kvinge, Pacific Northwest National Laboratory henry.kvinge@pnnl.gov
  Tegan Emerson, Pacific Northwest National Laboratory
  Tim Doster, Pacific Northwest National Lab
  Scott Mahan, Pacific Northwest National Laboratory
  Sarah McGuire, Michigan State University
  

  • 1:00 p.m.
    Uniform convergence guarantees for adversarially robust learning
    Rachel Morris*, North Carolina State University
    Ryan W. Murray, North Carolina State University
    (1203-68-44530)
  • 1:30 p.m.
    Convergence rates for deterministic generative diffusion models.
    Matt Jacobs*, UCSB
    (1203-35-44349)
  • 2:00 p.m.
    ReLU transformers and piecewise polynomials
    Zehua Lai*, University of Texas at Austin
    Lek-Heng Lim, University of Chicago
    Yucong Liu, Georgia Institute of Technology
    (1203-68-41124)
  • 2:30 p.m.
    POLICE: Provable Linear Constraint Enforcement for Deep Networks
    Randall Balestriero*, Brown University
    (1203-57-44632)
  • 3:00 p.m.
    Break
  • 3:30 p.m.
    Critical points of ReLU neural networks: Analytics and Empirics
    Marissa Masden*, University of Puget Sound
    (1203-57-42962)
  • 4:00 p.m.
    Oblique Randomized Decision Trees and Dimension Reduction
    Ricardo Baptista, Caltech
    Eliza O'Reilly*, Johns Hopkins University
    Yangxinyu Xie, The University of Texas at Austin
    (1203-60-43553)
  • 4:30 p.m.
    An interpretation for the role of depth in a deep neural network
    Thomas Chen, University of Texas at Austin
    Patricia Munoz Ewald*, University of Texas at Austin
    (1203-68-41213)