Paper Info
Title
Learning-assisted evolutionary search for scalable function optimization: LEM(ID3)
Abstract
Inspired originally by the Learnable Evolution Model(LEM), we investigate LEM(ID3), a hybrid of evolutionary search with ID3 decision tree learning. LEM(ID3) involves interleaved periods of learning and evolution, adopting the decision tree construction algorithm ID3 as the learning method, and a steady state EA as the evolution component. In the learning periods, ID3 is used to infer rules that attempt to identify `good' regions for genes, based on the values of one or more other genes. The rules are then used to guide the generation of new individuals. Without any preliminary parameter tuning, we evaluate LEM(ID3) on the test suite of 25 functions designed for the CEC 2005 special session on Real-Parameter Optimization. We describe the results, and in particular compare LEM(ID3) with the three most successful algorithms from the CEC 2005 competition; Sinha et al's K-PCX, and two versions of Auger and Hansen's CMA-ES. We find that LEM(ID3)'s performance is competitive with these algorithms, increasingly so as the problem dimensionality increases. In the case of 50-Dimensions, LEM(ID3) clearly records better overall performance on this function suite than the three comparative algorithms. Potential explanation for the trend in these results may lie in LEM(ID3)'s ability to repeatedly discover and be guided by a small number of relatively reliable, and perhaps nonlinear, inter-parameter relationships. Research-strength LEM(ID3) code is freely available at http://www.macs.hw.ac.uk/gls3/LEMID3/LEMID3.zip.
Year
DOI
Venue
2010
10.1109/CEC.2010.5586226
IEEE Congress on Evolutionary Computation
Keywords
Field
DocType
decision trees,evolutionary computation,learning (artificial intelligence),search problems,ID3 decision tree learning,LEM(ID3),decision tree construction algorithm,learning-assisted evolutionary search,real-parameter optimization,scalable function optimization
Test suite,Decision tree,Mathematical optimization,Algorithm design,Computer science,Learnable Evolution Model,Evolutionary computation,Curse of dimensionality,Artificial intelligence,ID3,Decision tree learning,Machine learning
Conference
ISBN
Citations 
PageRank 
978-1-4244-6909-3
2
0.36
References 
Authors
4
2
Name
Order
Citations
PageRank
Guleng Sheri120.36
David W. Corne22161152.00