DMCA
Automatic Feature Selection in Neuroevolution
BibTeX
@MISC{Computer_automaticfeature,
author = {Shimon Whitesondepartment Of Computer and Kenneth O. Stanleydepartment Of Computer},
title = {Automatic Feature Selection in Neuroevolution},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
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. Categories and Subject Descriptors I.2.6 [Artificial Intelligence]: Learning--Connectionism and neural nets General Terms Experimentation
Keyphrases
automatic feature selection regular neat appropriate set training data learning connectionism neat neuroevolution method abstract feature selection current machine previous effort initial experiment subject descriptor autonomous car neural net general term experimentation expensive meta-learning simulation demonstrate feature selection problem novel method feature selection task artificial intelligence feature selection
