mlsanityData to 7 Oct 2026

PatchAD: A Lightweight Patch-based MLP-Mixer for Time Series Anomaly Detection

arXiv 2401.09793 ↗First posted 18 Jan 2024cs.LGcited by 12

Not on the map. Why?

Abstract (arXiv)

Time series anomaly detection is a pivotal task in data analysis, yet it poses the challenge of discerning normal and abnormal patterns in label-deficient scenarios. While prior studies have largely employed reconstruction-based approaches, which limit the models' representational capacities. Moreover, existing deep learning-based methods are not sufficiently lightweight. Addressing these issues, we present PatchAD, our novel, highly efficient multiscale patch-based MLP-Mixer architecture that utilizes contrastive learning for representation extraction and anomaly detection. With its four distinct MLP Mixers and innovative dual project constraint module, PatchAD mitigates potential model degradation and offers a lightweight solution, requiring only 0.403M parameters. Its efficacy is demonstrated by state-of-the-art results across 8 datasets sourced from different application scenarios, outperforming over 30 comparative algorithms. PatchAD significantly improves the classical F1 score by 6.84%, the Aff-F1 score by 4.27%, and the V-ROC by 2.49%. Simultaneously, an in-depth analysis of the mechanisms underlying PatchAD has been conducted from both theoretical and experimental perspectives, validating the design motivations of the model. The code is publicly available at https://github.com/EmorZz1G/PatchAD.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 36 of its 94 references that are in the map. First posted on arXiv 18 Jan 2024.

Builds on

36 of 94 references.

References · 36

  1. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  2. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  3. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,590
  4. 2020Exploring Simple Siamese Representation Learningcited by 2,853
  5. 2021SimCSE: Simple Contrastive Learning of Sentence Embeddingscited by 2,168
  6. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 1,981
  7. 2021MLP-Mixer: An all-MLP Architecture for Visioncited by 1,441
  8. 2022TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysiscited by 1,030
  9. 2023One Fits All:Power General Time Series Analysis by Pretrained LMcited by 615
  10. 2021TS2Vec: Towards Universal Representation of Time Seriescited by 463
  11. 2018Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholdingcited by 407
  12. 2021Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancycited by 387
24 more
  1. 2021Time-Series Representation Learning via Temporal and Contextual Contrastingcited by 342
  2. 2022A Unified Model for Multi-class Anomaly Detectioncited by 267
  3. 2023TSMixer: An All-MLP Architecture for Time Series Forecastingcited by 232
  4. 2022TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Datacited by 211
  5. 2017A Multimodal Anomaly Detector for Robot-Assisted Feeding Using an LSTM-based Variational Autoencodercited by 209
  6. 2023Frequency-domain MLPs are More Effective Learners in Time Series Forecastingcited by 170
  7. 2020Current Time Series Anomaly Detection Benchmarks are Flawed and are Creating the Illusion of Progresscited by 148
  8. 2021Learning Graph Structures with Transformer for Multivariate Time Series Anomaly Detection in IoTcited by 104
  9. 2023DCdetector: Dual Attention Contrastive Representation Learning for Time Series Anomaly Detectioncited by 103
  10. 2022Deep Isolation Forest for Anomaly Detectioncited by 90
  11. 2019U-Time: A Fully Convolutional Network for Time Series Segmentation Applied to Sleep Stagingcited by 87
  12. 2022Local Evaluation of Time Series Anomaly Detection Algorithmscited by 79
  13. 2022How Does SimSiam Avoid Collapse Without Negative Samples? A Unified Understanding with Self-supervised Contrastive Learningcited by 65
  14. 2020Time Series Change Point Detection with Self-Supervised Contrastive Predictive Codingcited by 58
  15. 2023TimesURL: Self-supervised Contrastive Learning for Universal Time Series Representation Learningcited by 51
  16. 2021Do We Really Need Deep Learning Models for Time Series Forecasting?cited by 45
  17. 2021Neural Contextual Anomaly Detection for Time Seriescited by 42
  18. 2021Anomaly Detection of Time Series with Smoothness-Inducing Sequential Variational Auto-Encodercited by 39
  19. 2022Calibrated One-class Classification for Unsupervised Time Series Anomaly Detectioncited by 39
  20. 2017Collective Anomaly Detection based on Long Short Term Memory Recurrent Neural Networkcited by 28
  21. 2023PatchMixer: A Patch-Mixing Architecture for Long-Term Time Series Forecastingcited by 20
  22. 2023Nominality Score Conditioned Time Series Anomaly Detection by Point/Sequential Reconstructioncited by 17
  23. 2023When Model Meets New Normals: Test-time Adaptation for Unsupervised Time-series Anomaly Detectioncited by 17
  24. 2022Contrastive predictive coding for Anomaly Detection in Multi-variate Time Series Datacited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 3; 2025: 5; 2026: 4.

12 papers.

  1. 2024DACAD: Domain Adaptation Contrastive Learning for Anomaly Detection in Multivariate Time Seriescited by 7
  2. 2024A New Perspective on Time Series Anomaly Detection: Faster Patch-based Broad Learning Systemcited by 2
  3. 2025PatchTrAD: A Patch-Based Transformer focusing on Patch-Wise Reconstruction Error for Time Series Anomaly Detectioncited by 2
  4. 2026Anomalies in Multivariate Time Series Benchmarks Are Mostly Univariatecited by 2
  5. 2025Cluster-Aware Causal Mixer for Online Anomaly Detection in Multivariate Time Seriescited by 1
  6. 2025TriP-LLM: A Tri-Branch Patch-wise Large Language Model Framework for Time-Series Anomaly Detectioncited by 1
  7. 2026RAPT: Model-Predictive Out-of-Distribution Detection and Failure Diagnosis for Sim-to-Real Humanoid Deploymentcited by 1
  8. 2026TS-MLLM: A Multi-Modal Large Language Model-based Framework for Industrial Time-Series Big Data Analysiscited by 1
  9. 2024Anomalous Change Point Detection Using Probabilistic Predictive Codingcited by 0
  10. 2025CCE: Confidence-Consistency Evaluation for Time Series Anomaly Detectioncited by 0
  11. 2025PathFinder: Advancing Path Loss Prediction for Single-to-Multi-Transmitter Scenariocited by 0
  12. 2026PaAno+: Multiscale Encoding and Cross-Variable Attention for Time Series Anomaly Detectioncited by 0