Self-Trained One-class Classification for Unsupervised Anomaly Detection

06/11/2021
by   Jinsung Yoon, et al.
0

Anomaly detection (AD), separating anomalies from normal data, has various applications across domains, from manufacturing to healthcare. While most previous works have shown to be effective for cases with fully or partially labeled data, they are less practical for AD applications due to tedious data labeling processes. In this work, we focus on unsupervised AD problems whose entire training data are unlabeled and may contain both normal and anomalous samples. To tackle this problem, we build a robust one-class classification framework via data refinement. To refine the data accurately, we propose an ensemble of one-class classifiers, each of which is trained on a disjoint subset of training data. Moreover, we propose a self-training of deep representation one-class classifiers (STOC) that iteratively refines the data and deep representations. In experiments, we show the efficacy of our method for unsupervised anomaly detection on benchmarks from image and tabular data domains. For example, with a 10 method outperforms state-of-the-art one-class classification method by 6.3 AUC and 12.5 average precision.

READ FULL TEXT

page 9

page 13

research
02/13/2023

Unsupervised Deep One-Class Classification with Adaptive Threshold based on Training Dynamics

One-class classification has been a prevailing method in building deep a...
research
04/20/2021

What is Wrong with One-Class Anomaly Detection?

From a safety perspective, a machine learning method embedded in real-wo...
research
08/25/2023

A Generic Machine Learning Framework for Fully-Unsupervised Anomaly Detection with Contaminated Data

Anomaly detection (AD) tasks have been solved using machine learning alg...
research
10/14/2022

Learning image representations for anomaly detection: application to discovery of histological alterations in drug development

We present a system for anomaly detection in histopathological images. I...
research
06/06/2023

Efficient Anomaly Detection with Budget Annotation Using Semi-Supervised Residual Transformer

Anomaly Detection is challenging as usually only the normal samples are ...
research
01/17/2023

FewSOME: Few Shot Anomaly Detection

Recent years have seen considerable progress in the field of Anomaly Det...
research
03/15/2020

Self-trained Deep Ordinal Regression for End-to-End Video Anomaly Detection

Video anomaly detection is of critical practical importance to a variety...

Please sign up or login with your details

Forgot password? Click here to reset