AI Colloquium

RC Trust and the Department of Statistics at TU Dortmund welcome Professor Jin Xu to deliver a lecture on his latest work, “Bayesian Optimization via Exact Penalty” on 17 December. The talk takes place from 14:15 to 15:00 in room M/E 21 as part of the AI Colloquium and is open to anyone interested in how modern statistical tools help solve demanding optimization problems.

Professor Xu, a statistician at East China Normal University with a background in biostatistics and Bayesian methods, investigates how complex systems can be improved when simulations are costly and constraints are difficult to handle. His new approach – Exact Penalty Bayesian Optimization (EPBO) – combines statistical surrogate models with numerical optimization. This makes it possible to work efficiently even when starting from infeasible points or when equality constraints would normally slow down computation. EPBO aims to deliver reliable results under tight budgets, which is relevant from engineering design to applied research and decision-making.

The lecture offers insight for researchers exploring optimization methods, for students curious about practical applications of Gaussian processes, and for anyone interested in how data-driven approaches can support better solutions in science, technology and beyond.

About AI Colloquium

The AI Colloquium is a series of lectures dedicated to cutting-edge research in the field of machine learning and artificial intelligence, coorganized by the RC Trust, the Lamarr Institute for Machine Learning and Artificial Intelligence (Lamarr Institute) and the Center for Data Science & Simulation at TU Dortmund University (DoDas).