Jan Gmys
University of Mons
27 Papers
56 Citations
Jan Gmys is an academic researcher from University of Mons. The author has contributed to research in topics: Computer science & Branch and bound. The author has an hindex of 6, co-authored 21 publications. Previous affiliations of Jan Gmys include university of lille & French Institute for Research in Computer Science and Automation.
Chat about Author
Papers
A computationally efficient Branch-and-Bound algorithm for the permutation flow-shop scheduling problem
TL;DR: This work presents a new node decomposition scheme that combines dynamic branching and lower bound refinement strategies in a computationally efficient way and demonstrates that parallel tree search is a key ingredient for the resolution of large problem instances, as strong super-linear speedups can be observed.
76
A comparative study of high-productivity high-performance programming languages for parallel metaheuristics
Jan Gmys,Jan Gmys,Tiago Carneiro,Nouredine Melab,El-Ghazali Talbi,Daniel Tuyttens,Daniel Tuyttens +6 more
TL;DR: Three productivity-aware languages (Chapel, Julia, and Python) are compared in terms of performance, scalability and productivity, the first time such a comparison is performed in the context of parallel metaheuristics.
33
Parallel surrogate-assisted optimization: Batched Bayesian Neural Network-assisted GA versus q-EGO
Guillaume Briffoteaux,Guillaume Briffoteaux,Maxime Gobert,Maxime Gobert,Romain Ragonnet,Jan Gmys,Jan Gmys,Mohand Mezmaz,Nouredine Melab,Daniel Tuyttens +9 more
TL;DR: The study presented in this paper proves that parallel batched BNN-GA is a viable alternative to q-EGO approaches being more suitable for high-dimensional problems, parallelization impact, bigger data-bases and moderate search budgets.
32
A GPU-based Branch-and-Bound algorithm using IntegerVectorMatrix data structure
Jan Gmys,Mohand-Said Mezmaz,Nouredine Melab,Daniel Tuyttens +3 more
- 01 Nov 2016
TL;DR: This paper revisits the IVM-based B&B algorithm on the GPU, addressing the irregularity of the algorithm in terms of workload, memory access patterns and control flow, and focuses on reducing thread divergence by making a judicious choice for the mapping of threads onto the data.
Paradiseo: from a modular framework for evolutionary computation to the automated design of metaheuristics: 22 years of Paradiseo
Johann Dreo,Arnaud Liefooghe,Sébastien Verel,Marc Schoenauer,Juan J. Merelo,Alexandre Quemy,Benjamin Bouvier,Jan Gmys +7 more
- 07 Jul 2021
TL;DR: Paradiseo as mentioned in this paper is a C++ free software framework for the development of modular metaheuristics, which provides a highly modular architecture, a large set of components, speed of execution and automated algorithm design features.
23