Deep learning regularization in imbalanced data

Firuz Kamalov, Ho Hon Leung

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Deep neural networks are known to have a large number of parameters which can lead to overfitting. As a result various regularization methods designed to mitigate the model overfitting have become an indispensable part of many neural network architectures. However, it remains unclear which regularization methods are the most effective. In this paper, we examine the impact of regularization on neural network performance in the context of imbalanced data. We consider three main regularization approaches: L{1}, L{2}, and dropout regularization. Numerical experiments reveal that the L{1} regularization method can be an effective tool to prevent overfitting in neural network models for imbalanced data. Index Terms-regularization, neural networks, imbalanced data.

Original languageEnglish
Title of host publicationProceedings of the 2020 IEEE International Conference on Communications, Computing, Cybersecurity, and Informatics, CCCI 2020
EditorsMohammad S. Obaidat, Kuei-Fang Hsiao, Petros Nicopolitidis, Yu Guo, Daniel Cascado-Caballero
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728120355
DOIs
Publication statusPublished - Nov 3 2020
Event2020 IEEE International Conference on Communications, Computing, Cybersecurity, and Informatics, CCCI 2020 - Sharjah, United Arab Emirates
Duration: Nov 3 2020Nov 5 2020

Publication series

NameProceedings of the 2020 IEEE International Conference on Communications, Computing, Cybersecurity, and Informatics, CCCI 2020

Conference

Conference2020 IEEE International Conference on Communications, Computing, Cybersecurity, and Informatics, CCCI 2020
Country/TerritoryUnited Arab Emirates
CitySharjah
Period11/3/2011/5/20

Keywords

  • imbalanced data
  • neural networks
  • regularization

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems
  • Information Systems and Management
  • Safety, Risk, Reliability and Quality

Fingerprint

Dive into the research topics of 'Deep learning regularization in imbalanced data'. Together they form a unique fingerprint.

Cite this