Reinforcement Learning: Basic Concepts and Modern Applications in Traffic Management and Healthcare
Dist. Prof. Michael N. Katehakis
INFORMS (Institute for Operations Research and the Management Sciences) Fellow
Management Science and Information Systems Department, Rutgers University, NJ, USA
E-mail: [email protected]
Abstract: This talk introduces Reinforcement Learning (RL), surveying foundational concepts and modern applications in traffic management and healthcare. It begins with key RL principles and algorithms before exploring how RL optimizes traffic flow through adaptive signal control and smart routing. In healthcare, the focus shifts to RL's role in personalizing treatments and enhancing clinical trials. The presentation highlights RL's substantial potential in these fields, discussing both current implementations and fu ture possibilities. The talk concludes by discussing future directions for RL research and potential societal impacts of widespread RL adoption in traffic management.
Brief Biography of the Speaker:
Dr. Katehakis is a distinguished professor at Rutgers University and chair of the Department of Management Science & Information Systems. He holds a courtesy appointment in Rutgers' New Brunswick Department of Mathematics Graduate Faculty, and he is a member of DIMACS the Center for Discrete Mathematics and Theoretical Computer Science, he is a Primary Investigator of CDDA the Rutgers Center for Dynamic Data Analytics, and a member of RUTCOR, the Rutgers Center for Operations Research.
Much of his work has been on the interaction between optimization and statistical inference. Specific research interests include Stochastic Models, Dynamic Programming, Statistical Analysis and their application to Operations Management problems of pricing, production planning, inventory control, supply chains and scheduling. Many of these subjects are now known as Business Analytics a rapidly developing field at Google and at IBM.
He has co-authored many papers, with distinguished leaders in his field including: Cy Derman, Herbert E. Robbins, Sheldon M. Ross, Arthur F. Veinott Jr., Jerzy Filar, Uriel Rothblum, and Govindarajulu Z. with whom he won the 1992 Wolfowitz Prize for the paper "Dynamic allocation in survey sampling''. His work has been published in top journals and it has been funded by grants from the NSF and the AFOSR. Many of his Ph.D. students and their academic descendants are listed in The Mathematics Genealogy Project.
Professor Katehakis joined the Rutgers University faculty in 1989 after receiving his doctorate in Operations Research at Columbia University under the supervision of Cyrus Derman, and after being a faculty member at SUNY Stony Brook and at the Technical University of Crete. In addition, professor Katehakis was a member of the technical staff at the Operations Research Center of Bell - Laboratories, West Long Branch and a consultant at Brookhaven National Laboratory and he has held visiting appointments and taught at Columbia University, Stanford University and the National and Kapodistrian University of Athens, Greece.
Prof. Michael N. Katehakis is a Fellow of the Institute for Operations Research and the Management Sciences (INFORMS), an Elected Member of the International Statistical Institute (ISI) and a Senior Member of the Institute of Electrical and Electronics Engineers (IEEE).