Speaker: Pietro Consoli
School of Computer Science, University of Birmingham, UK
Title: Adaptive Crossover Operator Selection Based on Online Learning and Fitness Landscape Metrics
Date: May 14, 2015 5:00 PM BST
Self-adaptive mechanisms for the identification of the most suitable variation operator in evolutionary meta-heuristics rely almost exclusively on the measurement of the fitness of the offspring, which may not be sufficient to assess the optimality of an operator (e.g., in a landscape with a high degree of neutrality). In this talk, I will present a novel Adaptive Operator Selection mechanism which uses a set of four Fitness Landscape Analysis techniques and an online learning algorithm, Dynamic Weighted Majority, to provide more detailed information about the search space in order to better determine the most suitable crossover operator to use on a set of Capacitated Arc Routing Problem (CARP) instances. Extensive comparison with a state of the art approach has proved that this technique is able to produce comparable results on the set of benchmark problems and that, in some cases, it can outperform the results achieved adopting the fitness based approach.
This webinar is part of the IEEE CIS Junior Webinar Series.
After registering, you will receive a confirmation email containing information about joining the webinar.