Research

Paper

TESTING February 23, 2026

Unsupervised Anomaly Detection in NSL-KDD Using $β$-VAE: A Latent Space and Reconstruction Error Approach

Authors

Dylan Baptiste, Ramla Saddem, Alexandre Philippot, François Foyer

Abstract

As Operational Technology increasingly integrates with Information Technology, the need for Intrusion Detection Systems becomes more important. This paper explores an unsupervised approach to anomaly detection in network traffic using $β$-Variational Autoencoders on the NSL-KDD dataset. We investigate two methods: leveraging the latent space structure by measuring distances from test samples to the training data projections, and using the reconstruction error as a conventional anomaly detection metric. By comparing these approaches, we provide insights into their respective advantages and limitations in an unsupervised setting. Experimental results highlight the effectiveness of latent space exploitation for classification tasks.

Metadata

arXiv ID: 2602.19785
Provider: ARXIV
Primary Category: cs.LG
Published: 2026-02-23
Fetched: 2026-02-24 04:38

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.19785v1</id>\n    <title>Unsupervised Anomaly Detection in NSL-KDD Using $β$-VAE: A Latent Space and Reconstruction Error Approach</title>\n    <updated>2026-02-23T12:42:00Z</updated>\n    <link href='https://arxiv.org/abs/2602.19785v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.19785v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>As Operational Technology increasingly integrates with Information Technology, the need for Intrusion Detection Systems becomes more important. This paper explores an unsupervised approach to anomaly detection in network traffic using $β$-Variational Autoencoders on the NSL-KDD dataset. We investigate two methods: leveraging the latent space structure by measuring distances from test samples to the training data projections, and using the reconstruction error as a conventional anomaly detection metric. By comparing these approaches, we provide insights into their respective advantages and limitations in an unsupervised setting. Experimental results highlight the effectiveness of latent space exploitation for classification tasks.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.NE'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='stat.ML'/>\n    <published>2026-02-23T12:42:00Z</published>\n    <arxiv:primary_category term='cs.LG'/>\n    <arxiv:journal_ref>2025 15th France-Japan \\&amp; 13th Europe-Asia Congress on Mechatronics (MECATRONICS) / 23rd International Conference on Research and Education in Mechatronics (REM), Dec 2025, Saint-Ouen-sur-Seine, France. pp.1-6</arxiv:journal_ref>\n    <author>\n      <name>Dylan Baptiste</name>\n      <arxiv:affiliation>CRESTIC</arxiv:affiliation>\n    </author>\n    <author>\n      <name>Ramla Saddem</name>\n      <arxiv:affiliation>CRESTIC</arxiv:affiliation>\n    </author>\n    <author>\n      <name>Alexandre Philippot</name>\n      <arxiv:affiliation>CRESTIC</arxiv:affiliation>\n    </author>\n    <author>\n      <name>François Foyer</name>\n    </author>\n  </entry>"
}