Speaker: Mustafa Mısır, Associate Professor, Duke Kunshan University, China
Date: 26 October 2023
Time: 9:00 - 10:00pm (New Zealand Time, UTC+13)
Mustafa Misir is an Associate Professor of Data and Computational Science at Duke Kunshan University in China. He completed his Ph.D. in Computer Science at KU Leuven (Belgium) in 2012. After graduation, he worked as a postdoctoral researcher at INRIA Saclay - Universite Paris Sud XI (France), Singapore Management University (SMU) and University of Freiburg (Germany) respectively. Afterwards, he relocated to Nanjing University of Aeronautics and Astronautics (China) as a faculty member in the College of Computer Science and Technology. Prior to joining Duke Kunshan University, he was a faculty member in Computer Engineering at Istinye University (Turkiye). His main research interests include Automated Algorithm Design (Machine Learning + Algorithm Design) / Automated Machine Learning (AutoML) and Operations Research. He is the recipient of several prestigious academic awards and has published over 50 papers in various international conferences/journals.
Hyper-heuristics are problem-independent methods that are used to solve various instances of different problem domains. There have been effective hyper-heuristic designs in the literature that provide a certain level of generality in problem solving. Nonetheless, existing research indicates that no single hyper-heuristic performs consistently best in different problem-solving scenarios. Algorithm selection has primarily been investigated to address this issue, yet for problem-specific algorithms. The aim is to identify the (near) best algorithm(s) for each given problem instance. The idea here is to perform algorithm selection on selection hyper-heuristics, delivering problem-independent, cross-domain algorithm selection (cdAS). This talk will introduce cdAS and its empirical analysis on 9 combinatorial optimization problems, while also discussing hyper-heuristics and algorithm selection.
The Webinar went very successful, with 50+ participants.
The recorded video of the Webinar can be found here.
Back to the Webinar Series