(十二)Outlier Detection for Multidimensional Time Series using Deep Neural Networks

内含动画的PPT已经上传,可以在我上传的资源里看到,可免费下载

论文链接:Outlier Detection for Multidimensional Time Series using Deep Neural Networks

论文信息

  • 2018
  • IEEE
  • 多维时间序列+异常检测+时间序列的富集化+2DCNN-AE+LSTM-AE

一、论文概括

  1. 研究对象
  2. 目标
  3. 方法
  4. 结论

(十二)Outlier Detection for Multidimensional Time Series using Deep Neural Networks

二、相关引入工作

  1. 自编码器(AE)
  2. 引入本文方法
  3. 本文贡献

(十二)Outlier Detection for Multidimensional Time Series using Deep Neural Networks

(十二)Outlier Detection for Multidimensional Time Series using Deep Neural Networks

三、作者提出的方法

  1. 模型
  2. 时间序列的富集化(两步走)
  3. 基于子自编码器的深度神经网络
  4. 嵌入上下文信息
  5. “新的”架构
  6. 实验分析

(十二)Outlier Detection for Multidimensional Time Series using Deep Neural Networks

(十二)Outlier Detection for Multidimensional Time Series using Deep Neural Networks
(十二)Outlier Detection for Multidimensional Time Series using Deep Neural Networks

(十二)Outlier Detection for Multidimensional Time Series using Deep Neural Networks

(十二)Outlier Detection for Multidimensional Time Series using Deep Neural Networks

(十二)Outlier Detection for Multidimensional Time Series using Deep Neural Networks
(十二)Outlier Detection for Multidimensional Time Series using Deep Neural Networks

(十二)Outlier Detection for Multidimensional Time Series using Deep Neural Networks