Communication Dans Un Congrès Année : 2025

A Hybrid Autoencoder-Transformer Model for Detection of Attacks on Low Latency Services

Résumé

This paper addresses the problem of detection of attacks in computer networks. More precisely, we consider attacks on emerging low-latency services, which typically require a specific traffic management system. We present a simple yet very efficient hybrid method that takes advantage of both autoencoder and transformer models. The original method is compared with the current state-of-the-art on a large real-life dataset of network traffic to show the relevance of the proposed approach, especially for low false-positive rates. A quick ablation analysis shows that the efficiency of the method relies on the combinaison of the two methods jointly used in our hybrid model.

Fichier principal
Vignette du fichier
L4S_RC_AMLDS_HAL.pdf (606.84 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-05016498 , version 1 (02-04-2025)

Licence

Identifiants

  • HAL Id : hal-05016498 , version 1

Citer

Rémi Cogranne, Marius Letourneau, Guillaume Doyen. A Hybrid Autoencoder-Transformer Model for Detection of Attacks on Low Latency Services. 2025 IEEE International Conference on Advanced Machine Learning and Data Science (AMLDS 2025), IEEE, Jul 2025, Tokyo, Japan. ⟨hal-05016498⟩
490 Consultations
995 Téléchargements

Partager

  • More