Artificial intelligence and intelligent systems
V. Janko, M. Gams, B. Filipič, T. Šef, M. Luštrek, T. Tušar, A. Gradišek, E. Dovgan, M. Mlakar, J. Krivec,...
Computational intelligence is a study of stochastic search, optimization and learning methods, inspired by natural systems. Research in this area at the JSI Department of Intelligent Systems focuses on evolutionary computation and optimization. We study evolutionary algorithms for multiobjective optimization, their enhancement through parallel computing and surrogate models, constraint handling in multiobjective optimization, visualization of optimization results, and optimization algorithm benchmarking.
Our research findings and developed algorithms represent a basis for applications in diverse areas including:
V. Janko, M. Gams, B. Filipič, T. Šef, M. Luštrek, T. Tušar, A. Gradišek, E. Dovgan, M. Mlakar, J. Krivec,...
A. Vodopija (Supervised by B. Filipič) Within the doctoral research project, we aim at characterizing constrained continuous multiobjective optimization problems...
B. Filipič, T. Tušar, A. Vodopija, J. Cork This applied project deals with simulation-based optimization in electric motor design. The...
B. Filipič, T. Tušar, A. Vodopija, A. Andova
T. Tušar, A. Vodopija We collaborated with the Department of Knowledge Technologies and the Senso4s d.o.o. company in designing a...
B. Filipič, T. Tušar, A. Vodopija Scheduling of field work entails solving a demanding routing and scheduling problem. We assist...
B. Filipič, T. Tušar, A. Vodopija This is a bilateral, Slovenian-German project carried out together with the Institute for Data...
B. Filipič, E. Dovgan A project in collaboration with the MPT company
B. Filipič, E. Dovgan A project in collaboration with the Plamtex INT company
T. Tušar, E. Dovgan The goal of the TOPP (Transparent design of optimal underground connections for smart cities using artificial intelligence...
B. Filipič, T. Tušar In collaboration with the INEA company we developed optimization support for scheduling of flexible offers for...
B. Filipič. Zupančič, T. Tušar In this applied project, a simulation-based multiobjective optimization software was developed to support continuous casting...
T. Tušar, B. Naujoks The Job Market is a GECCO event (organized every year since 2017) where people offering jobs...
D. Brockhoff, T. Tušar In this tutorial, we discuss the past and future of benchmarking multiobjective optimizers. In particular, we...
B. Filipič, A. Vodopija This tutorial provides an overview of the state of the art in constraint handling in multiobjective...
B. Filipič, T. Tušar This tutorial provides an overview of representative methods used in multiobjective optimization for visualizing: (1) individual...
B. Filipič This tutorial presents the challenges faced and lessons learned in designing evolutionary algorithms for industrial optimization problems that...
B. Filipič, T. Tušar The 13th International Conference on Parallel Problem Solving from Nature, PPSN 2014, was co-organized by the...
Dimo Brockhoff, Anne Auger, Nikolaus Hansen, Tea Tušar
Quantitative performance assessment of multiobjective optimizers : the average runtime attainment function Inproceedings
In: Heike Trautmann (Ed.): Evolutionary multi-criterion optimization : 9th International Conference, EMO 2017, Münster, Germany, March 19-22, 2017 : proceedings, pp. str. 103-119, Springer, Cham, 2017.
@inproceedings{COBISS.SI-ID:30461735,
title = {Quantitative performance assessment of multiobjective optimizers : the average runtime attainment function},
author = {Dimo Brockhoff and Anne Auger and Nikolaus Hansen and Tea Tušar},
editor = {Heike Trautmann},
year = {2017},
date = {2017-01-01},
booktitle = {Evolutionary multi-criterion optimization : 9th International Conference, EMO 2017, Münster, Germany, March 19-22, 2017 : proceedings},
volume = {lncs 10173},
pages = {str. 103-119},
publisher = {Springer},
address = {Cham},
series = {Lecture notes in computer science, LNCS 10173},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Eric Medvet, Tea Tušar
The DU map : a visualization to gain insights into genotype-phenotype mapping and diversity Inproceedings
In: GECCO'17 companion : proceeding of the 2017 on Genetic and Evolutionary Computation Conference Companion, July 15-19, 2017 Berlin, Germany, pp. str. 1705-1712, ACM = Association for Computing Machinery, New York, 2017.
BibTeX | Links:
@inproceedings{COBISS.SI-ID:30637863,
title = {The DU map : a visualization to gain insights into genotype-phenotype mapping and diversity},
author = {Eric Medvet and Tea Tušar},
doi = {10.1145/3067695.3082554},
year = {2017},
date = {2017-01-01},
booktitle = {GECCO'17 companion : proceeding of the 2017 on Genetic and Evolutionary Computation Conference Companion, July 15-19, 2017 Berlin, Germany},
pages = {str. 1705-1712},
publisher = {ACM = Association for Computing Machinery},
address = {New York},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Tea Tušar, Nikolaus Hansen, Dimo Brockhoff
Anytime benchmarking of budget-dependent algorithms with the COCO platform Inproceedings
In: Mitja Luštrek, Rok Piltaver, Matjaž Gams (Ed.): Slovenska konferenca o umetni inteligenci : zbornik 20. mednarodne multikonference Informacijska družba - IS 2017, 9.-13. oktober 2017, [Ljubljana, Slovenia] : zvezek A = Slovenian Conference on Artificial Intelligence : proceedings of the 20th International Multiconference Information Society - IS 2017, 9th-13th October, 2017, Ljubljana, Slovenia : volume A, pp. str. 47-50, Institut Jožef Stefan, Ljubljana, 2017.
@inproceedings{COBISS.SI-ID:30856231,
title = {Anytime benchmarking of budget-dependent algorithms with the COCO platform},
author = {Tea Tušar and Nikolaus Hansen and Dimo Brockhoff},
editor = {Mitja Luštrek and Rok Piltaver and Matjaž Gams},
year = {2017},
date = {2017-01-01},
booktitle = {Slovenska konferenca o umetni inteligenci : zbornik 20. mednarodne multikonference Informacijska družba - IS 2017, 9.-13. oktober 2017, [Ljubljana, Slovenia] : zvezek A = Slovenian Conference on Artificial Intelligence : proceedings of the 20th International Multiconference Information Society - IS 2017, 9th-13th October, 2017, Ljubljana, Slovenia : volume A},
pages = {str. 47-50},
publisher = {Institut Jožef Stefan},
address = {Ljubljana},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Department of Intelligent Systems
Jožef Stefan Institute
Jamova cesta 39
SI-1000 Ljubljana, Slovenia
Tel: +386 1 477 3968