Machine Learning and Multi-scale Modeling
Modern machine learning has had remarkable success in all kinds of AI applications, and is also poised to change fundamentally the way we do physical modeling. In this talk, I will give an overview on some of the theoretical and practical issues that I consider most important in this exciting area.
Distance Geometry and Geometric Algebra for Protein Structure Determination
Homeostasis, Catastrophes, and Networks
Kinetics of particles with short-range interactions
Machine Learning at Facebook
Machine intelligence for processing big data sets is big business. A statistical mathematician's point of view has led to (1) effective large-scale principal component analysis and singular value decomposition, and (2) some theoretical foundations for convolutional networks (convolutional networks underpin the recent revolution in artificial intelligence).
The Fragile Families Challenge: A Scientific Mass Collaboration
Electrokinetic Control of Interfacial Instabilities
This talk will describe three examples of interfacial dynamics – viscous fingering, deionization shock propagation, and dendritic electrodeposition – whose stability can be controlled by electrokinetic phenomena in charged porous media, as evidenced by both theory and experiments. Potential applications include electrically enhanced oil recovery, water purification by shock electrodialysis, and energy storage with metal batteries.
Handling non-convexity in low-rank approaches for semidefinite programming
A semidefinite program (SDP) is an optimization problem where one seeks to minimize a linear function of a positive semidefinite matrix X under linear constraints. SDPs have generated sustained interest since the nineties owing to their powerful expressiveness. Standard algorithms solve SDPs in polynomial time, but they fail in practice for large problems.