Speaker: Kevin Tierney

Abstract: Quickly finding high quality solutions to optimization problems is essential for efficient logistics, production, disaster response, and many more applications. Recently, solution methods based on deep learning have become competitive with "traditional" Operations Research approaches to solving optimization problems. However, these techniques remain limited in the types of problems they can address and rarely outperform the state-of-the-art methods. In this talk, we will examine deep learning techniques for solving optimization problems and assess their strengths and weaknesses for managerial decision support. In doing so, we hope to provide some insight into whether deep learning is just the latest trend or if it can really achieve best-in-class performance.

Short Bio: Kevin Tierney is currently University Professor for Decision and Operation Technologies in the Faculty of Business Administration and Economics at Bielefeld University. His research focuses on the frontier of machine learning and combinatorial optimization, with the goal of solving optimization problems to near optimality fast and with low human input. His group's work on "Neural Large Neighborhood Search" was awarded the distinguished paper award at the European Conference of Artificial Intelligence in 2020, and his group has published multiple papers on the topic of deep learning for combinatorial optimization in top AI conferences. He was previously an assistant professor in the Department of Information Systems at Paderborn University. He holds a PhD from the IT University of Copenhagen, an Sc.M. in Computer Science from Brown University and a B.S. in Computer Science from Rochester Institute of Technology.