Tutorial on Modern Linkage Learning Techniques in Combinatorial Optimization

 

 

 

 

 

 

 

 

Back to tutorial overview

 

 

 

 

 

 

 

 

 

 

 

Organizers

 

 

Michal Witold Przewozniczek (main organizer)

 

 

Department of Computer Science, Wroclaw University of Science and Technology, Poland

 

 

Dr.  Michal Witold Przewozniczek received the M.Sc. and Ph.D. degrees in computer science in 2003 and 2012, respectively, from Wroclaw University of Science and Technology (WUST), Poland. He is an Assistant Professor with the Wroclaw University of Science and Technology in the Department of Computational Intelligence. In 2019, while associated with WUST, he also worked as a researcher at the Department of Computer Science, The University of York, England. Michal’s research concentrates on the application of evolutionary algorithms to practical problems. To obtain this objective, he investigates the linkage learning techniques, multi-population approaches, dynamic management of subpopulations, and others. This experience helped in solving hard multi-objective industrial problems as a team member at The University of York. Michal is also a co-owner of a software company. He invented and implemented in the industry many prototype systems.

 

 

 

 

 

Marcin Michal Komarnicki (co-organizer)

 

 

Department of Computer Science, Wroclaw University of Science and Technology, Poland

 

 

Marcin Michal Komarnicki received the B.S. and M.S. degrees in computer science from Wroclaw University of Science and Technology, Poland, in 2013 and 2014, respectively, where he is currently pursuing the Ph.D. degree. His current research interests include large-scale optimization, linkage learning techniques, and multi-population approaches. One of the main directions of Marcin’s research is proposing more effective linkage learning techniques for continuous search spaces.

Marcin is also a professional software developer. His professional work includes the applications of evolutionary computation into the production process planning in industry.

 

 

 

 

 

 

 

 

 

 

The most important research achievements of tutorial organizers that concern linkage learning and were published in 2020/21 are as follows.

 

 

 

Journal papers

 

 

 

 

1.       M.W. Przewozniczek, M. M. Komarnicki, “Empirical linkage learning,” IEEE Transactions on Evolutionary Computation,  vol. 24, no. 6, pp. 1097-1111, 2020.

2.       M. W. Przewozniczek, P. Dziurzanski, S. Zhao, L. S. Indrusiak, “Multi-Objective Parameter-less Population Pyramid for Solving Industrial Process Planning Problems,” Swarm and Evolutionary Computation, 2021 (in press)

3.       M. W. Przewoźniczek, “Subpopulation initialization driven by linkage learning for dealing with the Long-Way-To-Stuck effect,” Information Sciences, vol. 521, pp. 62-80, 2020.

4.       M. W. Przewozniczek, R. Datta, K. Walkowiak, M. Komarnicki, “Splitting the fitness and penalty factor for temporal diversity increase in practical problem solving,” Expert Systems With Applications, vol. 145, pp.1-11, 2020.

5.       M.W. Przewozniczek, R, Goścień, P. Lechowicz, K. Walkowiak, "Metaheuristic Algorithms with Solution Encoding Mixing for Effective Optimization of SDM Optical Networks," Engineering Applications of Artificial Intelligence, 2020 (in press).

 

 

 

Conference papers

 

 

 

1.       M.W. Przewozniczek, M. M. Komarnicki, B. Frej, "Direct linkage discovery with empirical linkage learning," in Proceedings of the 2021 Genetic and Evolutionary Computation Conference (GECCO ’21). ACM,  2021 (in press).

2.       M. W. Przewozniczek, M. M. Komarnicki, "Fitness caching - from a minor mechanism to major consequences in modern evolutionary computation," in Proceedings of the IEEE Congress on Evolutionary Computation (CEC), 2021 (in press).

3.       M. W. Przewozniczek, B. Frej, M. M. Komarnicki, “On measuring and improving the quality of linkage learning in modern evolutionary algorithms applied to solve partially additively separable problems,” in Proceedings of the 2020 Genetic and Evolutionary Computation Conference (GECCO ’20). ACM, pp. 742–750, 2020.

4.       M. M. Komarnicki, M. W. Przewozniczek, T. Durda, ”Comparative Mixing for DSMGA-II,” in Proceedings of the 2020 Genetic and Evolutionary Computation Conference (GECCO ’20), ACM, pp. 708–716, 2020.

5.       S. Wozniak, M. W. Przewozniczek, and M. M. Komarnicki, ”Parameter-less population pyramid for permutation-based problems,” in Proceedings of the Parallel Problem Solving from Nature (PPSN XVI), pp. 418-430, 2020.