I’m on the organizing committee for MDS26 and am participating in four sessions over the week.
Panel: Applied Mathematics and AI: A Two-Way Conversation Mon., Nov. 16, 4:30–5:30pm, Grand Ballroom ABCD
This panel looks at the relationship between applied mathematics and AI in both directions. It asks how mathematics can make AI more reliable, for example through learning from sparse or noisy data, physical constraints, uncertainty quantification, and guarantees. It also asks how AI is changing the way mathematicians formulate problems, write code, and develop proofs, and what students still need to learn for themselves. The discussion builds on the SIAM AI Task Force Report, The Role of Applied Mathematics in a New Era of Artificial Intelligence. I’m a panelist with Bert Debusschere (Sandia) and Deanna Needell (UCLA). Carol Woodward (LLNL) moderates and Suzanne Weekes (SIAM CEO) chairs.
Minisymposium Talk: Missing Data and Tensor Decompositions Thu., Nov. 19, 9:30–9:55am, Room 250B
I’ve worked on missing data in CP tensor decomposition for about twenty years. In this talk I survey the methods developed for the problem and the theoretical, computational, and practical difficulties it raises. The focus is on recovering the factors themselves rather than on tensor completion. I also discuss how assumptions such as smoothness can help, and close with open questions. My talk opens Part III of the minisymposium Statistical Modeling Meets Tensor Modeling, organized by Carlos Llosa, Daniel Dunlavy, and David Hong.
Panel: SIAM Publications Update – AI, SIMODS, and Accessibility Thu., Nov. 19, 11:45am–12:15pm, Grand Ballroom ABCD
As SIAM Vice President for Publications, I join David Gleich (Purdue), Editor-in-Chief of the SIAM Journal on Mathematics of Data Science (SIMODS), for an update on SIAM Publications. Topics include AI and research integrity, news from SIMODS, and the new Joint Math Society Author Guidelines on Accessibility. Kivmars Bowling (SIAM) chairs. Readers, authors, editors, and instructors are all welcome to bring questions and feedback.
Session Chair: Invited Plenary by Rachel Ward, All Hands on Deck! Applied Mathematics in the Era of Ready-made Proofs Fri., Nov. 20, 8:15–9:00am, Grand Ballroom ABCD
I’m chairing Rachel Ward’s (UT Austin) plenary. She argues that AI systems may soon answer any mathematical question that existing theory can settle. That makes the applied mathematician’s main job deciding which questions are worth asking, and it opens the door to bigger, more interdisciplinary research. She will also talk about organizing the First Proof Project, an independent assessment of AI capabilities in research mathematics.