Unravel Anomalies: An End-to-end Seasonal-Trend Decomposition Approach for Time Series Anomaly Detection

Zhenwei Zhang, Ruiqi Wang, Ran Ding, Yuantao Gu
9/30/2023
cs.LGcs.AI

Official code: zhangzw16/TADNet (17 stars, MIT, last commit 2024-04-18).

Abstract

Traditional Time-series Anomaly Detection (TAD) methods often struggle with the composite nature of complex time-series data and a diverse array of anomalies. We introduce TADNet, an end-to-end TAD model that leverages Seasonal-Trend Decomposition to link various types of anomalies to specific decomposition components, thereby simplifying the analysis of complex time-series and enhancing detection performance. Our training methodology, which includes pre-training on a synthetic dataset followed by fine-tuning, strikes a balance between effective decomposition and precise anomaly detection. Experimental validation on real-world datasets confirms TADNet's state-of-the-art performance across a diverse range of anomalies.

View on arXivView PDF

Code Implementations(1)

zhangzw16/TADNetOfficial100%
MIT

Code for ICASSP 2024 paper "Unravel Anomalies: An End-to-end Seasonal-Trend Decomposition Approach for Time Series Anomaly Detection".

172PythonDec 14, 20232 years agoMIT

Similar Papers

An Improved Time Series Anomaly Detection by Applying Structural Similarity

87% match2025

THEMIS: Unlocking Pretrained Knowledge with Foundation Model Embeddings for Anomaly Detection in Time Series

86% match2025

Possibility for Proactive Anomaly Detection

85% match2025

Towards a General Time Series Anomaly Detector with Adaptive Bottlenecks and Dual Adversarial Decoders

85% match1 repo262026

Dive into Time-Series Anomaly Detection: A Decade Review

85% match2024

OneShotSTL: One-Shot Seasonal-Trend Decomposition For Online Time Series Anomaly Detection And Forecasting

84% match1 repo422023

Cite this paper

@article{zhang2023unravel,
  title  = {Unravel Anomalies: An End-to-end Seasonal-Trend Decomposition Approach for Time Series Anomaly Detection},
  author = {Zhenwei Zhang and Ruiqi Wang and Ran Ding and Yuantao Gu},
  year   = {2023},
  eprint = {2310.00268},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2310.00268v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2310.00268.svg)](https://codeofpaper.com/paper/2310.00268)

Discussion