Publication detail

Automated Neural Network Structure Design for Efficient Anomaly Identification

HOLASOVÁ, E. FUJDIAK, R. BLAŽEK, P. BOHAČÍK, A.

Original Title

Automated Neural Network Structure Design for Efficient Anomaly Identification

Type

conference paper

Language

English

Original Abstract

The creation of suitable and efficient tools for anomaly detection constitutes a crucial aspect of security, applicable not only to industrial networks but also to cyber-physical systems. This article elucidates a framework designed to automate the selection of an optimal deep neural network architecture, thereby expediting the creation and implementation of neural network-based tools. The framework presented here enables a rapid design of an Artificial Neural Network structure without necessitating user intervention. Its efficacy has been showcased through experimentation with the publicly accessible HAI dataset, yielding an accuracy of approximately 0.94 after 10 epochs. Subsequently, a second scenario was performed where a total of 5456 models were generated and trained, with an average time of approximately 9.95 seconds per model.

Keywords

Artificial Neural Network, Industrial Networks, Neural Network, Neural Network Design, Neural Network Structure Design, Security, Structure Optimization

Authors

HOLASOVÁ, E.; FUJDIAK, R.; BLAŽEK, P.; BOHAČÍK, A.

Released

3. 12. 2023

ISBN

979-8-4007-0796-4

Book

ICCNS 2023 Proceedings

Pages from

1

Pages to

7

Pages count

7

BibTex

@inproceedings{BUT185768,
  author="Eva {Holasová} and Radek {Fujdiak} and Petr {Blažek} and Antonín {Bohačík}",
  title="Automated Neural Network Structure Design for Efficient Anomaly Identification",
  booktitle="ICCNS 2023 Proceedings",
  year="2023",
  pages="1--7",
  isbn="979-8-4007-0796-4"
}