Bio of Sebastian Pokutta - Interactive Optimization and Learning

Artificial Intelligence, Optimization, and Machine Learning. Professor @TUBerlin, Vice President @ZuseInstitute, Chair @MATH+, Chair @MODAL
Sebastian Pokutta is a mathematician and computer scientist, the Vice President of the Zuse Institute Berlin (ZIB) and a Professor of Mathematics at TU Berlin with a research focus on Artificial Intelligence and Optimization. He is also the Executive Chair of the Cluster of Excellence MATH+, a Chair of the Research Campus MODAL, and a member of ELLIS (European Laboratory for Learning and Intelligent Systems). Having received both his diploma and Ph.D. in mathematics from the University of Duisburg-Essen in Germany, Pokutta was a postdoctoral researcher and visiting lecturer at MIT, worked for IBM ILOG, and various consulting companies. Prior to joining ZIB and TU Berlin, he was the David M. McKenney Family Associate Professor in the School of Industrial and Systems Engineering and an Associate Director of the Machine Learning @ GT Center at the Georgia Institute of Technology as well as a Professor at the University of Erlangen-Nürnberg. Sebastian received the Gödel Prize in 2023, a STOC Test of Time award (10 years) in 2022, the David M. McKenney Family Early Career Professorship in 2016, an NSF CAREER Award in 2015, the Coca-Cola Early Career Professorship in 2014, the outstanding thesis award of the University of Duisburg-Essen in 2006, as well as various Best Paper awards.
Pokutta’s research is situated at the intersection of Artificial Intelligence and Optimization, combining Machine Learning with continuous and discrete optimization techniques, alongside the complexity theory of Extended Formulations. A particular focus is the development and analysis of Frank-Wolfe (conditional gradient) methods due to their versatility in constrained optimization, structured learning, and network compression. More recently, his research group has focused on AI4Science and AI4Math, working on agentic systems for scientific discovery and mathematical reasoning. This includes the development of evolutionary coding agents, automated constraint-handler generation for mixed-integer programming, and the integration of Large Language Models with formal proof verification. This is complemented by the use of deep learning and attention-based architectures to solve problems in computational algebra, such as transformer oracles for border basis algorithms and neural sum-of-squares certificates. Additional research directions include the interface of AI and human co-creativity, modeling multi-agent student simulations, temporal persona stability, and functional analogs of intrinsic motivation. Pokutta has also worked on diverse applied optimization problems across supply chain logistics, manufacturing, and quantitative finance.