Presenter Bio:

R. Srikant is a Grainger Distinguished Chair in Engineering, co-Director of the C3.ai Digital Transformation Institute and a Professor in the Department of Electrical and Computer Engineering and the
Coordinated Science Lab at the University of Illinois, Urbana-Champaign. His research interests include applied probability, machine learning and communication networks. He is the recipient of the 2015 INFOCOM Achievement Award, the 2019 IEEE Koji Kobayashi Computers and Communications Award and the 2021 ACM SIGMETRICS Achievement Award. He has also received several Best Paper awards including the 2015 INFOCOM Best Paper Award, the 2017 Applied Probability Society Best Publication Award, and the 2017 WiOpt Best Paper award. He was the Editor-in-Chief of the IEEE/ACM Transactions on Networking from 2013-2017 and is currently the Stochastic Models Area Chair for Mathematics of Operations Research.

Abstract:

The first part of the talk will provide some motivation for the use of central limit theorem to design algorithms for machine learning and graph sampling. Then, we will provide an introduction to Stein’s method which can be used to upper bound the Wasserstein distance between the distribution of a scaled sum of random variables and a Gaussian distribution. Finally, we will present some new results on nonasymptotic CLTs for vector-valued martingales and vector-valued functions of Markov chains, and discuss an application to machine learning. The first part of the talk will draw upon several papers to provide motivation and an introduction to Stein’s method while the latter part of the talk will be based on this paper: https://arxiv.org/pdf/2401.15719.