| Title | Understanding Differential Evolution |
| Speaker | Dr. Ferrante Neri |
| Chair | Swagatam Das |
Abstract
Differential Evolution (DE) is a popular, efficient, and versatile optimiser that displays a great potential on continuous domains. Due to its performance and simplicity DE and its variations have been applied in various fields of engineering and applied sciences. On the other hand, over the last two decades computer scientists proposed many DE variants that enhance upon the performance of DE.
This keynote will describe DE mechanism, and working principle, focusing on strong and weak point of the search logic. It will be discussed what makes the success of DE and why still it is characterized by a wide margin of improvement. The main DE modifications will be analysed according to a top-down strategy in order to understand what are their common features and thus the operators that guarantee the DE enhancements. The ultimate scope of this keynote will thus be to understand how the DE scheme work and therefore to know how to use it and tailor it in order to tackle various optimisation problems.
Biography
Ferrante Neri received the Masters and Ph.D. degrees in electrical engineering from the Technical University of Bari, Italy, in 2002 and 2007, respectively, and the Ph.D. degree in computer science from the University of Jyväskylä, Finland, in 2007. In 2009, Dr. Neri has been appointed Academy Research Fellow with the Academy of Finland. Since 2010 he is Adjunct Professor in Computational Intelligence at the University of Jyväskylä. Currently he is a Reader in Computational Intelligence at the De Montfort University, UK. Dr. Neri is vice-chair of the IEEE Task Force on Differential Evolution and chair of the IEEE Task Force on Memetic Computing. His current research interests include memetic computing, differential evolution, noisy and large scale optimisation, and compact and parallel algorithms.
