Adversarially Learned Anomaly Detection

IEEE ICDM 2018
paper
code

研究動(dòng)機(jī)（主要解決的問(wèn)題）

1、developing effective methods for complex and high-dimensional data remains a challenge

對(duì)復(fù)雜的高維的數(shù)據(jù)難處理

2、The need to solve an optimization problem for every test example makes this method impractical on large datasets or for real-time applications

優(yōu)點(diǎn)：effective, but also efficient at test time.

框架方法

Loss & Anomaly Score

loss
$$\begin{array}{l}V\left(D_{xz},D_{xx},D_{zz},E,G\right)=V\left(D_{xz},E,G\right)+V\left(D_{xx},E,G\right)+V\left(D_{zz},E,G\right)\end{array}$$

Anomaly Score
$$A(x)={\Vert f_{xx}(x,x)?f_{xx}(x,G(E(x)))\Vert }_1$$

A(x) 表示D的置信度，樣本是都被很好的encoder或者reconstructed by generator。值越大表示越異常。

實(shí)驗(yàn)

數(shù)據(jù)集：

KDDCup99

Arrhythmia

參數(shù)設(shè)置:

KDDCup99 :20%的異常

Arrhythmia ：15%的異常

use 80% of the whole official dataset for training and keep the remaining 20% as our test set.

We further remove 25% from the training set for a validation set and discard anomalous samples from both training and validation sets (thus setting up a novelty detection task).

評(píng)價(jià)方法：

Precision, Recall, F1 score

baselines：

One Class Support Vector Machines (OC-SVM)

Support vector method for novelty detection 1999

Isolation Forests (IF)

Isolation forest 2008

Deep Structured Energy Based Models (DSEBM)

Deep structured energy based models for anomaly detection 2016

Deep Autoencoding Gaussian Mixture Model (DAGMM)

Deep autoencoding gaussian mixture model for unsupervised anomaly detection 2018

AnoGAN

Unsupervised anomaly detection with generative adversarial networks to guide marker discovery 2017

實(shí)驗(yàn)結(jié)果

總結(jié)

我們提出了一種基于GAN的異常檢測(cè)方法ALAD，它在訓(xùn)練期間從數(shù)據(jù)空間到潛在空間學(xué)習(xí)編碼器，使得它在測(cè)試時(shí)比單獨(dú)發(fā)布的GAN方法更有效。 此外，我們還采用了額外的鑒別器來(lái)改進(jìn)編碼器，以及已經(jīng)發(fā)現(xiàn)可以穩(wěn)定GAN訓(xùn)練的頻譜歸一化。

總結(jié)

以上是生活随笔為你收集整理的ALAD的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問(wèn)題。

如果覺(jué)得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。