Exploiting deep convolutional neural networks for a neural-based learning classifier system

Ji Yoon Kim, Sung-Bae Cho

Research output: Contribution to journalArticle

Abstract

Classification is a key factor in accuracy, simplicity, and expressiveness, and it is difficult to optimize all of these factors at the same time. The learning classifier system (LCS) is a suitable technique for addressing an adaptive classification problem. It is a combination of fast approximation and evolutionary optimization techniques. A neural-based learning classifier system (N-LCS) includes an architecture for maintaining expressiveness by incorporating neural networks into a supervised classifier system, which is also an LCS specializing in classification studies. In recent years, studies using deep artificial neural networks have been actively conducted. In particular, deep convolutional neural networks (CNN) provide a powerful representation in an extremely fundamental method and demonstrates the high performance in various domains. In this paper, we exploit various deep CNN architectures in convolutional neural-based learning classifier systems (CN-LCS) combining the CNN and LCS to explore the possibility of a CN-LCS. By using various CNNs as an action of a classifier in an N-LCS, better classification accuracy can be obtained and classifier can be optimized. Experimental results show that our models achieve the higher performance than N-LCS for database intrusion detection as well as two other datasets, and extract effective features from deep representation by projecting data samples learned by several deep CNN models into the feature space.

Original languageEnglish
Pages (from-to)61-70
Number of pages10
JournalNeurocomputing
Volume354
DOIs
Publication statusPublished - 2019 Aug 18

Fingerprint

Classifiers
Learning
Neural networks
Neural Networks (Computer)
Intrusion detection
Network architecture
Databases

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

@article{a05a0430b1a14dbc95ce3b94a23e76fe,
title = "Exploiting deep convolutional neural networks for a neural-based learning classifier system",
abstract = "Classification is a key factor in accuracy, simplicity, and expressiveness, and it is difficult to optimize all of these factors at the same time. The learning classifier system (LCS) is a suitable technique for addressing an adaptive classification problem. It is a combination of fast approximation and evolutionary optimization techniques. A neural-based learning classifier system (N-LCS) includes an architecture for maintaining expressiveness by incorporating neural networks into a supervised classifier system, which is also an LCS specializing in classification studies. In recent years, studies using deep artificial neural networks have been actively conducted. In particular, deep convolutional neural networks (CNN) provide a powerful representation in an extremely fundamental method and demonstrates the high performance in various domains. In this paper, we exploit various deep CNN architectures in convolutional neural-based learning classifier systems (CN-LCS) combining the CNN and LCS to explore the possibility of a CN-LCS. By using various CNNs as an action of a classifier in an N-LCS, better classification accuracy can be obtained and classifier can be optimized. Experimental results show that our models achieve the higher performance than N-LCS for database intrusion detection as well as two other datasets, and extract effective features from deep representation by projecting data samples learned by several deep CNN models into the feature space.",
author = "Kim, {Ji Yoon} and Sung-Bae Cho",
year = "2019",
month = "8",
day = "18",
doi = "10.1016/j.neucom.2018.05.137",
language = "English",
volume = "354",
pages = "61--70",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",

}

Exploiting deep convolutional neural networks for a neural-based learning classifier system. / Kim, Ji Yoon; Cho, Sung-Bae.

In: Neurocomputing, Vol. 354, 18.08.2019, p. 61-70.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Exploiting deep convolutional neural networks for a neural-based learning classifier system

AU - Kim, Ji Yoon

AU - Cho, Sung-Bae

PY - 2019/8/18

Y1 - 2019/8/18

N2 - Classification is a key factor in accuracy, simplicity, and expressiveness, and it is difficult to optimize all of these factors at the same time. The learning classifier system (LCS) is a suitable technique for addressing an adaptive classification problem. It is a combination of fast approximation and evolutionary optimization techniques. A neural-based learning classifier system (N-LCS) includes an architecture for maintaining expressiveness by incorporating neural networks into a supervised classifier system, which is also an LCS specializing in classification studies. In recent years, studies using deep artificial neural networks have been actively conducted. In particular, deep convolutional neural networks (CNN) provide a powerful representation in an extremely fundamental method and demonstrates the high performance in various domains. In this paper, we exploit various deep CNN architectures in convolutional neural-based learning classifier systems (CN-LCS) combining the CNN and LCS to explore the possibility of a CN-LCS. By using various CNNs as an action of a classifier in an N-LCS, better classification accuracy can be obtained and classifier can be optimized. Experimental results show that our models achieve the higher performance than N-LCS for database intrusion detection as well as two other datasets, and extract effective features from deep representation by projecting data samples learned by several deep CNN models into the feature space.

AB - Classification is a key factor in accuracy, simplicity, and expressiveness, and it is difficult to optimize all of these factors at the same time. The learning classifier system (LCS) is a suitable technique for addressing an adaptive classification problem. It is a combination of fast approximation and evolutionary optimization techniques. A neural-based learning classifier system (N-LCS) includes an architecture for maintaining expressiveness by incorporating neural networks into a supervised classifier system, which is also an LCS specializing in classification studies. In recent years, studies using deep artificial neural networks have been actively conducted. In particular, deep convolutional neural networks (CNN) provide a powerful representation in an extremely fundamental method and demonstrates the high performance in various domains. In this paper, we exploit various deep CNN architectures in convolutional neural-based learning classifier systems (CN-LCS) combining the CNN and LCS to explore the possibility of a CN-LCS. By using various CNNs as an action of a classifier in an N-LCS, better classification accuracy can be obtained and classifier can be optimized. Experimental results show that our models achieve the higher performance than N-LCS for database intrusion detection as well as two other datasets, and extract effective features from deep representation by projecting data samples learned by several deep CNN models into the feature space.

UR - http://www.scopus.com/inward/record.url?scp=85064927609&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064927609&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2018.05.137

DO - 10.1016/j.neucom.2018.05.137

M3 - Article

VL - 354

SP - 61

EP - 70

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -