Skip to main content

Synergy for Smart Multi-Objective Optimisation

Deliverables

Conference proceeding of the HPO in industry conference

This deliverable will be a collection of articles from a conference on high performance optimisation in industry to be held in Slovenia towards the end of the project.

Summer school course materials

This deliverable will comprise the materials of the summer school on smart multi-objective optimisation such as its syllabus and presentations. While the course materials will be prepared for the summer school attendees, they will also be made publicly available to all interested parties.

Potential complex optimisation problems in science and industry

Potential complex optimisation problems in science and industry: in this deliverable, we will report some strategic large scale applications in different domains where multi-objective optimisation and surrogate-assistant modelling are crucial. Explored domains will concern: health, smart factories, smart buildings & homes, smart cities & communications, and smart use of resources.

Report on enhancing high-performance computing

Report on enhancing high-performance computing: we will provide a tutorial and lecture on the use of large scale heterogeneous parallel architectures for solving complex optimization problems. The focus will be put on new trends of high-performance computing identified in TOP500 international ranking including multi-core processors and co-processors (GPU, MIC) and their combination.

Report on improving excellence in multi-objective optimisation

The reports on improving excellence in multi-objective optimisation describes the improvements with respect to the integration of USTL and CUAS expertise to improve the expertise of all partners involved. It will focus on the actions taken to merge large scale parallel optimisation algorithms and high quality surrogate models as well as the steps to make JSI a leading authority in the field(s).

Proceedings of the BIOMA conference

The proceedings will contain invited and regular papers accepted to and presented at the 8th International Conference on Bioinspired Optimization Methods and their Applications, BIOMA 2018, which is the main scientific meeting to be organized by the SYNERGY partners. The proceedings will be available in electronic and printed form.

Report on prepared key performance indicators

This deliverable will provide the key performance indicators on all publications in high impact journals in the relevant research fields..

Tutorial on surrogate-assisted modelling

The tutorial on surrogate-assisted modelling will compile the state of the art on the corresponding topic and provide best practice methods for different scenarios. It will review different modelling techniques and different procedures for the integration of surrogates in considered optimisation techniques like evolutionary algorithms.

Tutorial on parallel multi-objective optimisation

Tutorial on parallel multi-objective optimization: this tutorial will present the state of the art of multi-objective metaheuristics and evolutionary algorithms in terms of design and implementation. In terms of implementation, different parallel architectures will be taken into account including multi-core and many-core processors. Best practies will be also outlined considering the characteristics of applications.

Report on refining surrogate modelling

The report on refining surrogate modelling will describe the actions to be taken to transport CUAS know-how to JSI and the cooperation of JSI and CUAS within different events. These will spread from the organisation of events at different conferences to the dissemination of know-how to the areas from Slovenia’s Smart Specialisation Strategy.

Edited book on high-performance optimisation

Edited book on high-performance optimisation: this book will collect state of the art high-performance optimisation algorithms in which parallel computing and surrogate-assisted modeling are used. Different optimization algorithms, parallel architectures and applications will be tackled.

Project website

A project website (with public and private area).

Publications

Deep statistical comparison of meta-heuristic stochastic optimization algorithms

Author(s): Tome Eftimov, Peter Korošec, Barbara Koroušić Seljak
Published in: Proceedings of the Genetic and Evolutionary Computation Conference Companion on - GECCO '18, 2018, Page(s) 15-16
DOI: 10.1145/3205651.3208210

Comparing multi-objective optimization algorithms using an ensemble of quality indicators with deep statistical comparison approach

Author(s): Tome Eftimov, Peter Korosec, Barbara Korousic Seljak
Published in: 2017 IEEE Symposium Series on Computational Intelligence (SSCI), 2017, Page(s) 1-8
DOI: 10.1109/ssci.2017.8280910

The impact of statistics for benchmarking in evolutionary computation research

Author(s): Tome Eftimov, Peter Korošec
Published in: Proceedings of the Genetic and Evolutionary Computation Conference Companion on - GECCO '18, 2018, Page(s) 1329-1336
DOI: 10.1145/3205651.3208232

Evolution of electric motor design approaches : the domel case

Author(s): Gregor Papa, Gašper Petelin, and Peter Korošec
Published in: Mednarodna konferenca o visokozmogljivi optimizaciji v industriji, HPOI 2018 : zbornik 21. mednarodne multikonference Informacijska družba - IS 2018, 2018

From a production scheduling simulation to a digital twin

Author(s): Gregor Papa and Peter Korošec
Published in: Mednarodna konferenca o visokozmogljivi optimizaciji v industriji, HPOI 2018 : zbornik 21. mednarodne multikonference Informacijska družba - IS 2018, 2018

Anytime Benchmarking of Budget-Dependent Algorithms with the COCO Platform

Author(s): Tušar , Tea; Hansen , Nikolaus; Brockhoff , Dimo
Published in: https://hal.inria.fr/hal-01629087, Issue 1, 2017

Comparison of Multi-objective Approaches to the Real-World Production Scheduling

Author(s): Gregor Papa, Peter Korošec
Published in: Advances in Evolutionary and Deterministic Methods for Design, Optimization and Control in Engineering and Sciences, Issue 48, 2019, Page(s) 457-471
DOI: 10.1007/978-3-319-89988-6_27

Data-Driven Preference-Based Deep Statistical Ranking for Comparing Multi-objective Optimization Algorithms

Author(s): Tome Eftimov, Peter Korošec, Barbara Koroušić Seljak
Published in: Bioinspired Optimization Methods and Their Applications, Issue 10835, 2018, Page(s) 138-150
DOI: 10.1007/978-3-319-91641-5_12

Quantitative Performance Assessment of Multiobjective Optimizers: The Average Runtime Attainment Function

Author(s): Dimo Brockhoff, Anne Auger, Nikolaus Hansen, Tea Tušar
Published in: Evolutionary Multi-Criterion Optimization, Issue 10173, 2017, Page(s) 103-119
DOI: 10.1007/978-3-319-54157-0_8

A novel statistical approach for comparing meta-heuristic stochastic optimization algorithms according to the distribution of solutions in the search space

Author(s): Tome Eftimov, Peter Korošec
Published in: Information Sciences, Issue 489, 2019, Page(s) 255-273, ISSN 0020-0255
DOI: 10.1016/j.ins.2019.03.049

A study of overfitting in optimization of a manufacturing quality control procedure

Author(s): Tea Tu?ar, Klemen Gantar, Valentin Koblar, Bernard ?enko, Bogdan Filipi?
Published in: Applied Soft Computing, Issue 59, 2017, Page(s) 77-87, ISSN 1568-4946
DOI: 10.1016/j.asoc.2017.05.027

A GRASS GIS parallel module for radio-propagation predictions


Published in: ISSN 1365-8816
DOI: 10.1080/13658816.2013.879151

Multi-core implementation of the differential ant-stigmergy algorithm for numerical optimization

Author(s): Peter Korošec, Marian Vajteršic, Jurij Šilc, Rade Kutil
Published in: The Journal of Supercomputing, Issue 63/3, 2013, Page(s) 757-772, ISSN 0920-8542
DOI: 10.1007/s11227-012-0772-z

A multi-objective approach to the application of real-world production scheduling

Author(s): Peter Korošec, Uroš Bole, Gregor Papa
Published in: Expert Systems with Applications, Issue 40/15, 2013, Page(s) 5839-5853, ISSN 0957-4174
DOI: 10.1016/j.eswa.2013.05.035

Scalar vs. vector approach to bi-objective resource allocation in spatially distributed networks

Author(s): Bogdan Filipič, Risto Vesanen, Erkki Laitinen
Published in: International Journal of Innovative Computing and Applications, Issue 5/3, 2013, Page(s) 191, ISSN 1751-648X
DOI: 10.1504/IJICA.2013.055937

GP-DEMO: Differential Evolution for Multiobjective Optimization based on Gaussian Process models

Author(s): Miha Mlakar, Dejan Petelin, Tea Tušar, Bogdan Filipič
Published in: European Journal of Operational Research, Issue 243/2, 2015, Page(s) 347-361, ISSN 0377-2217
DOI: 10.1016/j.ejor.2014.04.011

Visualization of Pareto Front Approximations in Evolutionary Multiobjective Optimization: A Critical Review and the Prosection Method

Author(s): Tea Tusar, Bogdan Filipic
Published in: IEEE Transactions on Evolutionary Computation, Issue 19/2, 2015, Page(s) 225-245, ISSN 1089-778X
DOI: 10.1109/TEVC.2014.2313407

Comparing Solutions under Uncertainty in Multiobjective Optimization

Author(s): Miha Mlakar, Tea Tušar, Bogdan Filipič
Published in: Mathematical Problems in Engineering, Issue 2014, 2014, Page(s) 1-10, ISSN 1024-123X
DOI: 10.1155/2014/817964

Asynchronous Master-Slave Parallelization of Differential Evolution for Multi-Objective Optimization

Author(s): Matjaž Depolli, Roman Trobec, Bogdan Filipič
Published in: Evolutionary Computation, Issue 21/2, 2013, Page(s) 261-291, ISSN 1063-6560
DOI: 10.1162/EVCO_a_00076

Discovering driving strategies with a multiobjective optimization algorithm

Author(s): E. Dovgan, M. Javorski, T. Tušar, M. Gams, B. Filipič
Published in: Applied Soft Computing, Issue 16, 2014, Page(s) 50-62, ISSN 1568-4946
DOI: 10.1016/j.asoc.2013.11.014

Visualizing Exact and Approximated 3D Empirical Attainment Functions

Author(s): Tea Tušar, Bogdan Filipič
Published in: Mathematical Problems in Engineering, Issue 2014, 2014, Page(s) 1-18, ISSN 1024-123X
DOI: 10.1155/2014/569346

Comparing a multiobjective optimization algorithm for discovering driving strategies with humans

Author(s): E. Dovgan, M. Javorski, T. Tušar, M. Gams, B. Filipič
Published in: Expert Systems with Applications, Issue 40/7, 2013, Page(s) 2687-2695, ISSN 0957-4174
DOI: 10.1016/j.eswa.2012.11.006