zoukankan      html  css  js  c++  java
  • PP: Unsupervised anomaly detection via variational auto-encoder for seasonal KPIs in web applications

    Problem: unsupervised anomaly detection

    for seasonal KPIs in web applications. 

    Donut: an unsupervised anomaly detection algorithm based on VAE.

    Background:

    有的time series data have seasonal patterns occurring at regular intervals. 

    Data: KPI shapes: seasonal patterns and local variations, noises.  

    "abnormal": anomalies and missing points; detect missing points is straightforward.

    Existing methods suffer from: 这里面简直是胡说八道。

    1. the hassle麻烦 of algorithm picking
    2. parameter tuning
    3. heavy reliance on labels
    4. unsatisfying performance
    5. lack of theoretical foundations

    Methodology:

    VAE is not a sequential model!!!!!!!!!!!!!!!! thus they apply sliding windows. 

    在训练时,the anomalies and missing points in a testing window x can bring bias to the mapped z, and further make the reconstruction probability inaccurate. 

    如何避免anomalies and missing points对训练造成的bias:

    1. missing points. adopt the MCMC-based missing data imputation technique with the trained VAE. 即模拟出missing points的可能值,然后用可能值,代替missing points 的值。
    2. anomalies

    All the algorithms evaluated in this paper compute one anomaly score for each point. A threshold can be chosen to do the decision: if the score for a point is greater than the threshold, an alert should be triggered

  • 相关阅读:
    Spring Boot|Async
    SonarQube使用
    大数据|linux权限chmod和chown
    Spring Boot|监控-Actuator
    Docker|部署及简单使用
    Spring|IOC启动流程
    Spring|注解
    Spring|@Autowired与new的区别
    Spring|Spring MVC
    个人站点
  • 原文地址:https://www.cnblogs.com/dulun/p/12342402.html
Copyright © 2011-2022 走看看