Kas 2019
Seminer Duyurusu: "Selected Applications of Meta-heuristics" by Dr. Berna Kiraz

"Selected Applications of Meta-heuristics"
Konuşmacı : Dr. Berna Kiraz
Tarih : 11 Kasım 2019 (Pazartesi)
Saat : 13:30
Yer : Bilgisayar ve Bilişim Fakültesi, 4305

Many meta-heuristic approaches have been presented and successfully applied to solve different types of optimization problems. This talk will cover the applications of these methods to the following problems: (1) the multi-objective capacitated multiple allocation hub location problem, (2) the many-objective diet problem, (3) feature selection, and (4) inverse design of integrated photonic filter systems. First, I will present a meta-heuristic based on non-dominated sorting genetic algorithm (NSGA-II) for a multi-objective, capacitated, multiple allocation, and incomplete hub location problem. In this study, a mathematical model is presented for the problem. Then, we develop a multi-objective approach based on NSGA-II for decisions on the number of hubs, the locations of hubs, the allocations of spokes to hubs, the hub network design, and the routing of all demands. In the second study, we present a well-known many-objective evolutionary algorithm (NSGA-III) for diet problem. The problem is modeled as a many-objective and multi-constraint optimization problem. Based on the experimental results, our algorithm can make healthy meal recommendations which is applicable by different users. Third, I will introduce opposition based particle swarm optimization for feature subset selection problem. In the last study, the main goal is to obtain integrated photonic devices using inverse design methods. Integrated photonic devices constitute important parts of our modern communication networks that satisfy the increasing need for information access. To determine device (filter) geometry, analytical and heuristic methods is used in conjunction with electromagnetic simulations at multiple wavelengths. The talk will also include my future research plans.  

berna_kiraz__6Nov2019Berna Kiraz received her B.S. degree in Mathematics from Ege University in 2004, her M.S. degree in Computer Engineering from Marmara University in 2008, and her Ph.D. degree in Computer Engineering from Istanbul Technical University in 2014. She worked as a postdoctoral researcher at Computational Optimization & Innovation Lab at Michigan State University, East Lansing, MI, USA for one year. She was a research assistant at Marmara University between 2007 and 2017. She joined the Department of Computer Engineering at Fatih Sultan Mehmet Vakif University as an assistant professor in 2017. Her research interests include Evolutionary Computation, Meta-heuristics and Hyper-heuristics, Dynamic Single/Multi-objective Optimization Problems, Evolutionary Machine Learning.