http://www.titan.cs.unp.ac.za/~nelishiap/cec2015/image002.jpg

 

 

 

 

IEEE 2015 Congress on Evolutionary Computation (CEC 2015)

Special Session: Evolutionary Algorithms in Hyper-Heuristics(EAHH 2015)

Aims and Scope

 

Hyper-heuristics aim to provide a generalized solution for a particular problem domain or across different problem domains. This is achieved by employing methods, such as metaheuristics, to combine or generate low-level heuristics. The low-level heuristics can be constructive, i.e. are used to create a solution, or perturbative, in which case the heuristics improve a candidate solution. Based on the function of the hyper-heuristic and type of low-level heuristics, a hyper-heuristic can be categorized as selection constructive, selection perturbative, generation constructive or generation perturbative. Evolutionary algorithms have been employed by hyper-heuristics and have played a pivotal role in the generation, hybridization and selection of low-level heuristics. Evolutionary algorithm hyper-heuristics have successfully been applied to various domains including timetabling, vehicle routing, decision tree induction, packing problems, text classification and dynamic environments amongst others. In certain domains, e.g. timetabling, selection perturbative hyper-heuristics have proven to be more effective than direct exploration of the solution space by evolutionary algorithms. Evolutionary algorithms, specifically genetic programming and grammatical evolution, have primarily been employed by hyper-heuristics to generate low-level heuristics. Recent trends in this field include the use of hyper-heuristics for algorithm design and hybridization of methods. Algorithm design essentially involves determining the parameter values and methods to use, e.g. the method of selection and crossover and mutation probabilities in ant algorithms. Hybridization is achieved by means of a selection perturbative hyper-heuristic to hybridize different approaches to solve the problem at hand, e.g. different multi-objective evolutionary algorithms are low-level heuristics in a selection perturbative hyper-heuristic to solve multi-objective optimization problems. The aim of this special session is for researchers to present recent developments in the field thereby paving the way for future advancement.

 

Topics

 

The main topics include but are not limited to:

 

Applications of evolutionary algorithm hyper-heuristics

Theoretical aspects of evolutionary algorithm hyper-heuristics

Evolutionary algorithm hyper-heuristics for algorithm design

Evolutionary algorithm hyper-heuristics for the derivation of hybrid methods

Hybridization of evolutionary algorithm hyper-heuristics, i.e. the design of hyper-hyper-heuristics using evolutionary algorithms

Cross domain applications of evolutionary algorithm hyper-heuristics

Parallelization of evolutionary algorithm hyper-heuristics

 

 

Organizers

 

Nelishia Pillay,

University of KwaZulu-Natal,

E-mail: pillayn32@ukzn.ac.za

 

Rong Qu,

University of Nottingham,

E-mail: Rong.Qu@nottingham.ac.uk

 

Important Dates

 

Paper submission deadline: December 19, 2014 January 16, 2015

Paper acceptance notification: February 20, 2015

Final paper submission deadline: March 13, 2015

Early registration: March 13, 2015

 

Paper Submission

 

Special session papers are treated the same as regular papers and must be submitted via the CEC 2015 submission website. When submitting choose the "Evolutionary Algorithm in Hyper-Heuristics" special session from the "Main Research Topic" list.

 

 

Program Committee

 

TBA