A comparative study of an evolvability indicator and a predictor of expected performance for genetic programming

Published in Conferences Papers
  1. Leonardo Trujillo, Yuliana Mart\'ınez, Edgar Galván López and Pierrick Legrand. A Comparative Study of an Evolvability Indicator and a Predictor of Expected Performance for Genetic Programming. In Proceedings of the 14th Annual Conference Companion on Genetic and Evolutionary Computation. 2012, 1489–1490. URL, DOI BibTeX

    @inproceedings{Trujillo:2012:CSE:2330784.2331006,
    	author = "Trujillo, Leonardo and Mart\'{\i}nez, Yuliana and Galv\'{a}n L\'{o}pez, Edgar and Legrand, Pierrick",
    	title = "A Comparative Study of an Evolvability Indicator and a Predictor of Expected Performance for Genetic Programming",
    	booktitle = "Proceedings of the 14th Annual Conference Companion on Genetic and Evolutionary Computation",
    	series = "GECCO '12",
    	year = 2012,
    	isbn = "978-1-4503-1178-6",
    	location = "Philadelphia, Pennsylvania, USA",
    	pages = "1489--1490",
    	numpages = 2,
    	url = "http://doi.acm.org/10.1145/2330784.2331006",
    	doi = "10.1145/2330784.2331006",
    	acmid = 2331006,
    	publisher = "ACM",
    	address = "New York, NY, USA",
    	keywords = "classification, genetic programming, performance prediction"
    }
    
Abstract

An open question within Genetic Programming (GP) is how to characterize problemdifficulty. The goal is to develop predictive tools that estimate how difficult a problemis for GP to solve. Here we consider two groups of methods. We call the first group Evolvability Indicators (EI), measures that capture how amendable the fitness landscape is to a GP search. Examples of EIs are Fitness Distance Correlation (FDC) and Negative Slope Coefficient (NSC). The second group are Predictors of Expected Performance (PEP), models that take as input a set of descriptive attributes of a problem and predict the expected performance of GP. This paper compares an EI, the NSC, and a PEP model for a GP classifier. Results suggest that the EI does not correlate with the performance of the GP classifiers. Conversely, the PEP models show a high correlation with GP performance.

Published in
GECCO Companion '12 Proceedings of the fourteenth international conference on Genetic and evolutionary computation conference companion
Pages 1489-1490
http://dl.acm.org/citation.cfm?id=2331006
Date of conference
07 - 11 July 2012
ISBN
978-1-4503-1178-6
Feedback