Automatic Feature Selection in Neuroevolution. Shimon Whiteson,
Peter Stone, Kenneth O. Stanley,
Risto Miikkulainen, and Nate Kohl.
In GECCO 2005: Proceedings of the Genetic and Evolutionary Computation Conference, pp. 1225–1232, June 2005.
http://www.isgec.org/gecco-2005/
Feature selection is the process of finding the set of inputs to a machine learning algorithm that will yield the best performance. Developing a way to solve this problem automatically would make current machine learning methods much more useful. Previous efforts to automate feature selection rely on expensive meta-learning or are applicable only when labeled training data is available. This paper presents a novel method called FS-NEAT which extends the NEAT neuroevolution method to automatically determine an appropriate set of inputs for the networks it evolves. By learning the network's inputs, topology, and weights simultaneously, FS-NEAT addresses the feature selection problem without relying on meta-learning or labeled data. Initial experiments in an autonomous car racing simulation demonstrate that FS-NEAT can learn better and faster than regular NEAT. In addition, the networks it evolves are smaller and require fewer inputs. Furthermore, FS-NEAT's performance remains robust even as the feature selection task it faces is made increasingly difficult.
@InProceedings{whiteson:gecco05,
author = "Shimon Whiteson and Peter Stone and Kenneth O. Stanley and Risto Miikkulainen and Nate Kohl",
title = "Automatic Feature Selection in Neuroevolution",
booktitle = "GECCO 2005: Proceedings of the Genetic and Evolutionary Computation Conference",
month = "June",
year = "2005",
pages = "1225--1232",
abstract = {
Feature selection is the process of finding the set of inputs to a
machine learning algorithm that will yield the best performance.
Developing a way to solve this problem automatically would make
current machine learning methods much more useful. Previous efforts
to automate feature selection rely on expensive meta-learning or are
applicable only when labeled training data is available. This paper
presents a novel method called FS-NEAT which extends the NEAT
neuroevolution method to automatically determine an appropriate set
of inputs for the networks it evolves. By learning the network's
inputs, topology, and weights simultaneously, FS-NEAT addresses the
feature selection problem without relying on meta-learning or
labeled data. Initial experiments in an autonomous car racing
simulation demonstrate that FS-NEAT can learn better and faster than
regular NEAT. In addition, the networks it evolves are smaller and
require fewer inputs. Furthermore, FS-NEAT's performance remains
robust even as the feature selection task it faces is made
increasingly difficult.
}
wwwnote = {<a href="http://www.isgec.org/gecco-2005/">http://www.isgec.org/gecco-2005/</a>},
bib2html_pubtype = {Refereed Conference},
bib2html_rescat = {Machine Learning}
}
Generated by bib2html.pl (written by Patrick Riley ) on Sat Aug 22, 2009 06:50:12