Special sessions

Metaheuristics Optimization

Organized by Vida Vukašinović and Jurij Šilc

International Symposium on Operations Research in Slovenia (SOR 2017)

Bled, 27-29 September 2017

Metaheuristics can be viewed as strategies that navigate the search for near-optimal solutions in empirical optimization. They are designed to efficiently explore the search space under limited computational resources. Metaheuristic algorithms use stochastics components and are typically not problem-specific. Examples include simulated annealing, genetic algorithms, differential evolution, ant colony optimization, bee algorithms, particle swarm optimization, tabu search, harmony search and many others. The field is fast growing, with notable success in solving real-world problems, but with the lack of theoretical insight and often insufficient elaboration of empirical results.

This special session addresses theoretical and empirical studies of metaheuristic optimization algorithms. Papers dealing with their utilization in operations research are particularly welcome. We look for high-quality research papers that represent novel contributions to the field. The scope of the special session covers, but is not limited to the following topics:

  • Theoretical analysis and algorithm models
  • Hybrid / parallel / distributed metaheuristics
  • Multi- and many-objective optimization
  • Simulation-based and surrogate-based optimization
  • Comparative studies
  • Real-world applications

EMO – Evolutionary Multiobjective Optimization

Organized by Tea Tušar and Carlos M. Fonseca

Genetic and Evolutionary Computation Conference (GECCO 2017)

Berlin, 17-18 July 2017

In many real-world applications, several objective functions have to be optimized simultaneously, leading to a multi-objective optimization problem (MOP) for which an ideal solution seldom exists. Rather, MOPs typically admit multiple compromise solutions representing different trade-offs among the objectives. Due to their applicability to a wide range of MOPs, including black-box optimization problems, evolutionary algorithms for multiobjective optimization have given rise to an important and very active research area, known as Evolutionary Multiobjective Optimization (EMO). No continuity or differentiability assumptions are required by EMO algorithms, and problem characteristics such as nonlinearity, multimodality and stochasticity can be handled as well. Furthermore, preference information provided by a decision maker can be used to deliver a finite-size approximation to the solution set (the so-called Pareto-optimal set) in a single optimization run.

Metaheuristics and Machine Learning

Organized by El-Ghazali Talbi, Nouredine Melab, and Peter Korošec

Metaheuristics International Conference (MIC 2017)

Barcelona, 7 July 2017

The session provides an opportunity to the international research community in optimization and learning to discuss recent research results and to develop new ideas and collaborations in a friendly and relaxed atmosphere. The session welcomes presentations that cover any aspects of optimization and learning research such as new high-impact applications, parameter tuning, 4th industrial revolution, new research challenges, hybridization issues, optimization-simulation, meta-modeling, high-performance and exascale computing, surrogate modeling, multi-objective optimization, optimization for machine learning, machine learning for optimization.

Multiobjective Optimization with Surrogate Models

Organized by Bogdan Filipič, Thomas Bartz-Beielstein, and Carlos A. Coello

IEEE World Congress on Computational Intelligence (WCCI 2016)

Vancouver, 29 July 2016

Many real-world optimization problems involve multiple, often conflicting objectives and rely on computationally expensive simulations to assess these objectives. Such multiobjective optimization problems can be solved more efficiently if the simulations are partly replaced by accurate surrogate models. Surrogate models, also known as response surface models or meta-models, are data driven models built to simulate the processes or devices that are subject to optimization. They are used when more precise models, such as those based on the finite element method or computational fluid dynamics, spend too much time and resources. While surrogate models allow for fast simulation and assessment of the optimization objectives, they also represent an additional source of impreciseness. In multiobjective optimization, this may constitute a particular challenge when comparing candidate solutions. The aim of this special session is to bring together researchers and practitioners working with surrogate-based multiobjective optimization algorithms to present recent achievements in the field and discuss directions for further work.

The scope of the special session covers, but is not limited to the following topics:

  • State-of-the-art in multiobjective optimization with surrogate models
  • Theoretical aspects of surrogate-assisted multiobjective optimization
  • Novel surrogate-based multiobjective optimization algorithms
  • Comparative studies in multiobjective optimization with surrogates
  • Benchmark problems and performance measures for multiobjective optimization with surrogates
  • Real-world applications of multiobjective optimization using surrogate

Coming events

  • Summer School
    27. 08. 2018 - 31. 08. 2018

    SYNERGY Summer School on Efficient Multi-Objective OptimisationLjubl

  • HPOI 2018
    08. 10. 2018

    International Conference on High-Performance Optimisation in Industry

  • META 2018
    27. 10. 2018 - 31. 10. 2018

    7th International Conference on Metaheuristics and Nature Inspired Computing