Description
Efficient utilization of resources as well as a general understanding about interdependencies between related processes is essential for companies in order to offer products and services in a competitive and sustainable manner. Therefore, systems modelling, analysis and optimization is a key-technology in manifold domains to solve complex practical problems. The presentation will include theoretical and practical showcases in two different application domains of metaheuristic optimization in general and evolutionary computation in special: First, several showcases will be presented about data based system analysis and optimization mainly in the field of steel producing industry as well as in automotive domains. Second, evolutionary algorithms based machine learning approaches will be shown with the aim to perform white-box modelling of nonlinear systems with the aim to detect interpretable models. Algorithmic strategies will be discussed that are able to integrate a-priory knowledge from the domain; especially already known laws from natural science. Also attention is given to support the uniqueness of models which are mainly discovered by symbolic regression techniques using enhanced evolutionary algorithms for the search in the space of hypotheses. The presentation will cover theoretical aspects as well as real world examples demonstrating how the open source framework HeuristicLab https://dev.heuristiclab.com/ can be used for modeling, optimization and machine learning tasks for concrete challenges in the domain of production, logistics and systems research. The potential of a generic framework for algorithm design, analysis and application will be one of the major inputs of the talk| Period | 17 Jul 2018 |
|---|---|
| Event title | Genetic and Evolutionary Computation Conference (GECCO 2018) |
| Event type | Conference |
| Location | Kyoto, Japan, JapanShow on map |