The Value (and Pitfalls) of Proof
Persi Diaconis (Stanford University)
Colloquium
Friday, November 1, 2024, 3:30 pm
Abstract
The computer is taking over. AND, in many branches of applied mathematics and statistics we hear 'why bother to prove theorems? simulations and numerical approximation are easier and better for real problems'. NOT SO FAST; I will present a collection of real world examples where the computer lies and thousands of hours of simulations give persistently wrong results. On the other hand, we theorem provers can prove useless, irrelevant things that choke off research and really do miss the interesting stuff. There is room for both of us, and there are interesting things to think about.
Bio
I am a mathematician and statistician working in probability, combinatorics, and group theory with a focus on applications to statistics and scientific computing. A specialty is rates of convergence of Markov chains. I am currently interested in trying to adapt the many mathematical developments to say something useful to practitioners in large real-world simulations.
This talk is viewable on our YouTube channel.
This event is co-sponsored by the Paul G. Allen School of Computer Science & Engineering, Statistics, and Mathematics.