Creative Commons Attribution 4.0 International license
This report documents the programme and outcomes of Dagstuhl Seminar 25451, "Bayesian Optimisation", held from November 2–7, 2025. The seminar brought together 39 international experts from machine learning, optimisation, statistics, and engineering to discuss recent advances, open challenges, and emerging research directions in Bayesian optimisation. The programme comprised plenary talks spanning foundational issues, benchmarking, and the growing interaction between Bayesian optimisation and generative AI, alongside focused working groups on key thematic areas. Beyond technical discussions, the seminar placed strong emphasis on community building and worked towards establishing best practices. This report summarises the plenary contributions, the outcomes of the working groups, and additional community-driven activities, including a collection of practical "tricks of the trade" and exploratory benchmarking exercises.
@Article{branke_et_al:DagRep.15.11.1,
author = {Branke, J\"{u}rgen and Hutter, Frank and Pedrielli, Giulia and Poloczek, Matthias and Papenmeier, Leonard},
title = {{Bayesian Optimisation (Dagstuhl Seminar 25451)}},
pages = {1--66},
journal = {Dagstuhl Reports},
ISSN = {2192-5283},
year = {2026},
volume = {15},
number = {11},
editor = {Branke, J\"{u}rgen and Hutter, Frank and Pedrielli, Giulia and Poloczek, Matthias and Papenmeier, Leonard},
publisher = {Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
address = {Dagstuhl, Germany},
URL = {https://drops.dagstuhl.de/entities/document/10.4230/DagRep.15.11.1},
URN = {urn:nbn:de:0030-drops-274940},
doi = {10.4230/DagRep.15.11.1},
annote = {Keywords: AutoML, Bayesian optimization, Benchmarking, Gaussian processes}
}