Speaker: Paul J. Atzberger

 

Title: An Introduction to Machine Learning: Foundations and Applications

 

In this talk we will survey recent approaches in machine learning, their mathematical foundations, and practical computational methods. Recent advances in measurement and scientific simulation are resulting in a plethora of data. This presents a number of challenges both for the design of high-throughput experiments/simulations and for the interpretation of the resulting large sets of data. Advances in combining approaches, such as Bayesian statistics, with large-scale scientific computation are resulted in new methodologies and algorithms for inferring information from such data.  We will give a survey of this area and practical methods drawing on results from stochastic analysis.  We will also discuss a few example applications.