Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Dec 5, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

860

A Spatiotemporal Deep Learning Approach for Unsupervised Anomaly Detection in Cloud Systems.

Zilong He, Pengfei Chen, Xiaoyun Li

    IEEE Transactions on Neural Networks and Learning Systems
    |October 16, 2020
    PubMed
    Summary

    This study introduces TopoMAD, a novel unsupervised anomaly detection model for cloud systems. TopoMAD effectively handles contaminated data and models complex spatiotemporal dependencies, improving system performance monitoring.

    Related Concept Videos

    You might also read

    Related Articles

    Articles linked to this work by shared authors, journal, and citation graph.

    Sort by
    Same author

    Scalable Manufacturing of Amino Acid-Based Piezoelectric Biocrystal Films.

    Small (Weinheim an der Bergstrasse, Germany)·2026
    Same author

    Diuranacyclobutadiene: Evidence of Electron Delocalization in the Actinide-Carbon Ring.

    Journal of the American Chemical Society·2026
    Same author

    Pharmacokinetics and anti-tumor efficacy of transarterial chemoembolization using small-particle lenvatinib-loaded beads in a rabbit VX2 liver tumor model.

    Quantitative imaging in medicine and surgery·2026
    Same author

    Performance Optimization of Liquid-Solid Nanogenerators With Fluorinated Alkyl Self-Assembled Monolayers.

    Small (Weinheim an der Bergstrasse, Germany)·2026
    Same author

    miR-20a-5p alleviates diabetic retinopathy by targeting TXNIP to inhibit NLRP3 inflammasome-dependent pyroptosis.

    Experimental eye research·2026
    Same author

    Factors Influencing Transition Readiness in Adolescents with Systemic Lupus Erythematosus: A Qualitative Study.

    Patient preference and adherence·2026

    Area of Science:

    • Cloud Computing
    • Machine Learning
    • System Performance Monitoring

    Background:

    • Anomaly detection is crucial for cloud system health.
    • Data-driven methods are prevalent but struggle with unlabeled, contaminated data.
    • Modeling complex spatiotemporal dependencies in large-scale systems is challenging.

    Purpose of the Study:

    • To develop a robust unsupervised anomaly detection model for cloud systems.
    • To effectively handle contaminated training data.
    • To model spatial and temporal dependencies in system metrics.

    Main Methods:

    • Proposed TopoMAD, a stochastic seq2seq model.
    • Utilized system topology to organize component metrics.
    • Employed graph neural networks for spatial feature extraction.

    More Related Videos

    Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
    08:20

    Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

    Published on: October 27, 2023

    2.2K

    Related Experiment Videos

    Last Updated: Dec 5, 2025

    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
    03:31

    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

    Published on: December 15, 2023

    860
    Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
    08:20

    Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

    Published on: October 27, 2023

    2.2K
  • Used long short-term memory networks for temporal feature extraction.
  • Developed the model using a variational auto-encoder for robustness.

    • Main Results:

    • TopoMAD demonstrated robust performance on contaminated data.
    • The model effectively captured spatiotemporal dependencies.
    • Validated on real-world data from big data and microservice systems.
    • Outperformed state-of-the-art anomaly detection methods.

    Conclusions:

    • TopoMAD offers a robust solution for unsupervised anomaly detection in complex cloud environments.
    • The model's ability to handle contaminated data and model spatiotemporal relationships is a significant advancement.
    • TopoMAD enhances the reliability and performance monitoring of cloud systems.