Upcoming Events
Towards unlocking the mystery of adversarial fragility of neural networks
Friday, May 2, 2025 3:30pm to 4:20pm
Speaker: Weiyu Xu, Dept. of Electrical and Computer Engineering
Mathematics Faculty Colloquium - Yangbo Ye
Thursday, May 8, 2025 3:30pm to 4:20pm
Title: Number theory and cryptography
Abstract: In this presentation Professor Ye will survey his work and progress in number theory since his last colloquium talk in the department. Topics include the Riemann zeta function, automorphic L-functions, their functoriality, upper bounds, and zero statistics, computational number theory, and its applications to cyberspace security or insecurity.
Short Bio: Professor Yangbo Ye is a professor of Mathematics with expertise in number theory and...
Past Events
Constrained Optimization Methods for Machine Learning with Fairness Constraints
Friday, February 21, 2025 3:30pm to 4:20pm
Speaker: Qihang Lin (Dept. of Business Analytics)
Evolution of Stress Response and Adhesin Gene Family in Pathogenic Yeasts
Friday, February 14, 2025 3:30pm to 4:20pm
Speaker: Bin He, Biology Dept.
Structural Models for Vascular Tissues
Friday, February 7, 2025 3:30pm to 4:20pm
Speaker: Jia Lu, Dept. Mechanical Engineering
Models of Mitochondrial Fission from ODE to PDE to DDE
Friday, January 31, 2025 3:30pm to 4:20pm
Speaker: Colleen Mitchell, Dept. of Mathematics
Time integration methods for systems with constraints
Friday, January 24, 2025 3:30pm to 4:20pm
Speaker: Laurent O. Jay, Department of Mathematics
Mathematics Faculty Colloquium - Xueyu Zhu; University of Iowa Department of Mathematics
Thursday, December 5, 2024 3:30pm to 4:20pm
Title: Recent Advancement of Scientific Machine Learning
Abstract: Machine learning has revolutionized scientific computing, offering unprecedented computational efficiency, flexibility, and applicability to real-world challenges. However, traditional machine learning approaches often overlook the rich insights provided by existing physical laws or mathematical properties. This talk explores the latest advancements in AI techniques that respect existing physical laws or mathematical properties...
A Machine Learning Approach to Analysis of Daily Vocal Function and Vocal Behavior of Individuals with Phonotraumatic Vocal Hyperfunction
Wednesday, September 25, 2024 4:00pm to 5:00pm
In-person (WJSHC 210) and via Zoom (https://uiowa.zoom.us/j/92047907158?pwd=35J1P4oAYnVfabrLo7xdtBFhrX22GC.1 )
Title:
A Machine Learning Approach to Analysis of Daily Vocal Function and Vocal Behavior of Individuals with Phonotraumatic Vocal Hyperfunction
Abstract:
Voice disorders affect approximately 8% of U.S. adults at a given point in time, with a lifetime prevalence of approximately 30% for adults. Many voice disorders are considered behavioral in nature and are associated with vocal...
Multigrid Methods in Space and Time for Extreme-scale Scientific Computing
Saturday, September 7, 2024 10:15am to 11:00am
Speaker: Dr. Rob Falgout, Center for Applied Scientific Computing (CASC) at Lawrence Livermore National Laboratory (LLNL)
Abstract: Multigrid methods play a key role in large-scale scientific simulation because they are among the fastest and most scalable approaches for solving systems of equations. They are widely used to solve the sparse spatial systems that arise in these simulations, and they have been shown to scale efficiently on today’s supercomputers. For time-dependent simulations...
Computational Math and Science Research at LLNL
Friday, September 6, 2024 4:30pm to 5:20pm
Speaker: Dr. Rob Falgout, Center for Applied Scientific Computing (CASC) at Lawrence Livermore National Laboratory (LLNL)
Math Colloquium - Speaker: Dr. Rob Falgout, Lawrence Livermore National Lab
Thursday, September 5, 2024 3:30pm to 4:20pm
Title: Parallel-in-Time Solution of Systems of Linear and Nonlinear Hyperbolic PDEs
Abstract: The multigrid reduction in time (MGRIT) method is a parallel multigrid-in-time solver designed to be as non-intrusive as possible and take advantage of existing simulation codes and techniques. This has worked well for parabolic equations, but parallel-in-time methods for advection-dominated hyperbolic problems have proven difficult to develop. In previous work, we demonstrated the effectiveness of a...