Research
Prof.dr. Peter A.N. Bosman is the group leader of the Evolutionary Intelligence research group at the Centrum Wiskunde & Informatica (CWI) (Center for Mathematics and Computer Science) located in Amsterdam, the Netherlands. He further has a part-time professor (in Dutch: deeltijdhoogleraar) position at Delft University of Technology in the Algorithmics group of the Department of Software Technology in the Faculty of Electrical Engineering, Mathematics, and Computer Science. Prof.dr. Bosman was formerly affiliated with Utrecht University, where he also obtained his M.Sc. and Ph.D. degrees in Computer Science.
Prof.dr. Bosman’s research is firmly rooted in Artificial Intelligence (AI), and revolves around the design and application of Evolutionary Algorithms (EAs) and combinations thereof with Machine Learning (ML) techniques. Key focus areas include
- model-based EAs, with a focus on the Gene‑pool Optimal Mixing Evolutionary Algorithm (GOMEA) family, co‑founded by prof.dr. Bosman. This includes variants for discrete, continuous, mixed, permutation‑based, and multi‑objective optimization, as well as GP‑GOMEA for symbolic regression. GOMEA has been used in multiple domains, including for one medical application now in clinical use.
- eXplainable AI (XAI) for multimodal data, with a focus on the development of the MultiFIX framework, enabling interpretable analysis of complex medical datasets and supporting a range of clinical research applications,
- multi-objective optimization and learning, where contributions include unique insights into the balance between proximity and diversity in evolutionary multi-objective optimization, how gradients in multi-objective optimization are themselves multi-objective, and how concepts from evolutionary computation could be used to achieve true multi-objective (deep) ML,
- high‑performance computing, with a focus on modern, highly parallel GPU‑based implementations (of GOMEA) that are essential for scaling to real‑world, computationally demanding applications.
Prof.dr. Bosman furthermore collaborates closely with domain experts to address real‑world problems requiring optimization and machine learning, often with a multi‑objective character. These collaborations lead to practically relevant problem formulations and novel algorithmic developments, increasing the likelihood of real‑world adoption. A major application area is the medical domain, where his work has been translated into clinical practice (e.g., prostate cancer brachytherapy at Amsterdam UMC). Additional application areas include logistics, (smart) energy systems, and materials design.
Prof.dr. Bosman has (co‑)authored numerous publications (for an overview, see Google Scholar), including 9 best‑paper award winning works, and 11 additional best-paper nominations. His work has also earned two silver Humies awards for human‑competitive results in medical applications of evolutionary algorithms. He currently serves as chair of SIGEVO, the ACM special interest group on Genetic and Evolutionary Computation, and previously held multiple leadership roles within the organization. He is an active program committee member for major conferences and journals in evolutionary computation and related fields. He has served as general chair, track chair, local chair, and organizer of workshops and tutorials at GECCO, the flagship conference in the field.
Finally, the (co-)acquired research grant funding by Prof.dr. Bosman totals over €13M, which includes funding from the Dutch research council, the Dutch cancer society, the Dutch children cancer-free foundation, and the European Innovation Council. Together, these grants support(ed) various scientific research positions (including 28 Ph.D. students and 7 postdocs, as well as various radiation therapy technologists, and scientific programmers), and various high-performance computing hardware.
