Dr. Jakob Bossek

Machine Learning and Optimisation

Academic Councillor

Office Address:
Fürstenallee 11
33102 Paderborn
Room:
FU.307

About Jakob Bossek

Jakob earned his bachelor's degree in Statistics and a diploma in Computer Science from TU Dortmund University in 2013 and 2014, respectively. In 2018, he successfully completed his doctoral studies in Information Systems at the University of Münster, Germany. Following this, he held PostDoc positions at both the University of Münster, Germany, and the University of Adelaide, Australia.

From April 2022 to November 2023, Jakob served as an Akademischer Rat (assistant professor) at the Chair for AI Methodology at RWTH Aachen. Presently, he holds the position of Akademischer Rat at the Chair for Machine Learning and Optimisation (MALEO) in the computer science department of Paderborn University.

Curriculum Vitae

04/2022 - 11/2023: Academic Councillor at the Department of Computer Science (Chair for AI Methodology), RWTH Aachen University, Germany

04/2021 - 03/2022: Academic Councillor at the Department of Information Systems (Chair for Statistics and Optimisation), University of Münster, Germany

09/2020 - 03/2021: PostDoc at the Department of Information Systems (Chair for Statistics and Optimization), University of Münster, Germany

10/2019 - 09/2020: PostDoc Researcher at the School of Computer Science, The University of Adelaide, Australia in the Optimisation and Logistics group of Prof. Dr. Frank Neumann

02/2015 - 09/2019: Research Associate (PhD student until Nov. 2018; later PostDoc) at the Department of Information Systems (Chair for Statistics and Optimization), University of Münster, Germany

11/2018: Doctoral degree in Information Systems (Dr. rer. pol.) at the University of Münster, Germany

09/2005 - 11/2014: Studying Computer Science with minor Statistics at the TU-Dortmund University (degree: Diplom; M.Sc. equivalent)

09/2008 - 05/2013: Studying Statistics with minor Computer Science at the TU-Dortmund University (degree: B.Sc.)

If you want truly to understand something, try to change it. - Kurt Levin

Research

Research Interests

His research interests span a wide spectrum, encompassing AutoAI methods, with a specific focus on algorithm selection and algorithm configuration. However, his primary research concentration lies in both empirical and theoretical aspects of bio-inspired problem-solving, particularly in the realm of evolutionary algorithms for NP-hard combinatorial optimization problems. Currently, his research endeavors aim to deepen the understanding of evolutionary algorithms through time complexity analysis in particular in the field of Quality Diversity (QD) and Evolutionary Diversity Optimisation (EDO) for combinatorial optimisation problems.

  • Multi-objective optimisation
  • Combinatorial optimisation
  • Heuristic optimisation
  • Evolutionary computation methods
  • Theory of randomised search heuristics
  • AutoAI methods (in particular algorithm selection and algorithm configuration)
  • Machine learning
  • Data analysis

Publications

Latest Publications

Using Reinforcement Learning for Per-Instance Algorithm Configuration on the TSP

M. Seiler, J. Rook, J. Heins, O.L. Preuß, J. Bossek, H. Trautmann, in: 2023 IEEE Symposium Series on Computational Intelligence (SSCI), IEEE, 2024.


On the Impact of Basic Mutation Operators and Populations within Evolutionary Algorithms for the Dynamic Weighted Traveling Salesperson Problem

J. Bossek, A. Neumann, F. Neumann, in: Proceedings of the Genetic and Evolutionary Computation Conference, Association for Computing Machinery, New York, NY, USA, 2023, pp. 248–256.


Runtime Analysis of Quality Diversity Algorithms

J. Bossek, D. Sudholt, in: Proceedings of the Genetic and Evolutionary Computation Conference, Association for Computing Machinery, New York, NY, USA, 2023, pp. 1546–1554.


Generating Diverse and Discriminatory Knapsack Instances by Searching for Novelty in Variable Dimensions of Feature-Space

A. Marrero, E. Segredo, E. Hart, J. Bossek, A. Neumann, in: Proceedings of the Genetic} and Evolutionary Computation Conference, Association for Computing Machinery, New York, NY, USA, 2023, pp. 312–320.


Do Additional Target Points Speed Up Evolutionary Algorithms?

J. Bossek, D. Sudholt, Theoretical Computer Science (2023) 113757.


Show all publications

Teaching


Current Courses

  • Unsupervised Learning and Evolutionary Optimisation Using R (in English)
  • Programmierung 1