Hans-Georg Beyer

Hans-Georg Beyer is a Research Professor for Operations Research, Modeling and Simulation at the Vorarlberg University of Applied Sciences in Dornbirn, Austria, where he leads research at the Research Center Business Informatics (formerly Center Process and Product Engineering). He is also a Co-opted Member of the Faculty of Engineering, Computer Science, and Psychology at Ulm University, Germany. Previously, he held positions at the Chair of Systems Analysis at the Department of Computer Science of the University of Dortmund, Germany, including as a Professor of Computer Science.

Dr. Beyer's research focuses on Evolutionary Computation, particularly Evolution Strategies and Genetic Algorithms, with significant contributions to theoretical analysis and practical applications. His work spans Computational Intelligence, Natural Computing, and various optimization techniques including noisy and constrained optimization. He has made substantial advancements in Matrix Adaptation Evolution Strategies (MA-ES), developing algorithms that efficiently handle large-scale and constrained optimization problems by learning correlations between decision variables.

His recent publications demonstrate a clear trend toward analyzing Evolution Strategies in highly multimodal landscapes, with particular focus on the Rastrigin and Griewank functions as benchmark test cases. Beyer's research has evolved from theoretical foundations to practical implementations, with a strong emphasis on understanding population sizing, self-adaptation mechanisms, and convergence properties in complex optimization scenarios. His work bridges theoretical analysis with practical algorithm design, making significant contributions to the field of evolutionary computation.

• Editor-in-Chief of Journal Evolutionary Computation (MIT-Press, 2010-2016)
• Associate Editor of IEEE Transactions on Evolutionary Computation (1997-2021)
• Member of VDI/VDE GMA Fachausschuss 5.14 Computational Intelligence
• Member of ACM Special Interest Group for Genetic and Evolutionary Computation
• Author of the Springer book "The Theory of Evolution Strategies"

Beyer has led several significant research projects including the FWF project "Evolutionary Global Optimization in Highly Multimodal Landscapes" (P 33702) and the FWF project "Evolution Strategies for Constrained Optimization." His research group has developed practical implementations of Evolution Strategies, including fast MA-ES with O(N^2) complexity, Meta MA-ES (Bi-Population MA-ES), and Limited Memory MA-ES (LM-MA-ES) for large-scale optimization problems. He regularly contributes tutorials at major conferences including GECCO and WCCI, sharing his expertise on Matrix Adaptation Evolution Strategies with the broader research community.