JoLEA
The twenty-sixth lecture takes place on the 25th of March 2026 at 4:30 PM (CET), virtually.
The zoom link to attend the lecture is: https://cwi-nl-zoom.zoom.us/j/81717018152?pwd=zV034FI6JsuqNHugY4YbdFfIUaKkNE.1&jst=2 The lecture will be held by Prof. Dr. Olaf Mersmann, Hochschule des Bundes für öffentliche Verwaltung, Germany.
The long road to automatic algorithm selection for black-box optimization problems
The talk will give a biased overview of the tools we, the BBOB community, have developed over the last decade or so to help us choose suitable algorithms or algorithm configurations for continuous black-box optimization problems. Some of the things I will discuss are: ranking, exploratory landscape analysis, problem similarity, real-world vs. analytic problems and how all of this ties together. Along the way, there will be a short tangent on the No Free Lunch theorem and its relevance or irrelevance to algorithm selection in the real world. To round it all off, I will present a superficially simple black-box problem with a cheap evaluation function and colorful results. Using the example, I hope to motivate you, that it is high time to reconsider some of our research goals and change your perspective on how difficult a “black-box” problem is.
About the speaker
Olaf Mersmann is a Professor at Hochschule des Bundes in Germany. He received his BSc, MSc and PhD in Statistics from TU Dortmund. His research interests include using statistical and machine learning methods on large benchmark databases to gain insight into the structure of the algorithm choice problem.
