PatchTrAD: A Patch-Based Transformer focusing on Patch-Wise Reconstruction Error for Time Series Anomaly Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Vilhes, Samy-Melwan, Gasso, Gilles, Alaya, Mokhtar Z
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909644835782656
author Vilhes, Samy-Melwan
Gasso, Gilles
Alaya, Mokhtar Z
author_facet Vilhes, Samy-Melwan
Gasso, Gilles
Alaya, Mokhtar Z
contents Time series anomaly detection (TSAD) focuses on identifying whether observations in streaming data deviate significantly from normal patterns. With the prevalence of connected devices, anomaly detection on time series has become paramount, as it enables real-time monitoring and early detection of irregular behaviors across various application domains. In this work, we introduce PatchTrAD, a Patch-based Transformer model for time series anomaly detection. Our approach leverages a Transformer encoder along with the use of patches under a reconstructionbased framework for anomaly detection. Empirical evaluations on multiple benchmark datasets show that PatchTrAD is on par, in terms of detection performance, with state-of-the-art deep learning models for anomaly detection while being time efficient during inference.
format Preprint
id arxiv_https___arxiv_org_abs_2504_08827
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PatchTrAD: A Patch-Based Transformer focusing on Patch-Wise Reconstruction Error for Time Series Anomaly Detection
Vilhes, Samy-Melwan
Gasso, Gilles
Alaya, Mokhtar Z
Machine Learning
Artificial Intelligence
Time series anomaly detection (TSAD) focuses on identifying whether observations in streaming data deviate significantly from normal patterns. With the prevalence of connected devices, anomaly detection on time series has become paramount, as it enables real-time monitoring and early detection of irregular behaviors across various application domains. In this work, we introduce PatchTrAD, a Patch-based Transformer model for time series anomaly detection. Our approach leverages a Transformer encoder along with the use of patches under a reconstructionbased framework for anomaly detection. Empirical evaluations on multiple benchmark datasets show that PatchTrAD is on par, in terms of detection performance, with state-of-the-art deep learning models for anomaly detection while being time efficient during inference.
title PatchTrAD: A Patch-Based Transformer focusing on Patch-Wise Reconstruction Error for Time Series Anomaly Detection
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2504.08827