Explainable deep few-shot anomaly detection with deviation networks

Existing anomaly detection paradigms overwhelmingly focus on training detection models using exclusively normal data or unlabeled data (mostly normal samples). One notorious issue with these approaches is that they are weak in discriminating anomalies from normal samples due to the lack of the knowl...

Full description

Saved in:
Bibliographic Details
Main Authors: PANG, Guansong, DING, Choubo, SHEN, Chunhua, HENGEL, Anton Van Den
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2021
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/7036
https://ink.library.smu.edu.sg/context/sis_research/article/8039/viewcontent/2108.00462.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-8039
record_format dspace
spelling sg-smu-ink.sis_research-80392022-03-24T07:13:24Z Explainable deep few-shot anomaly detection with deviation networks PANG, Guansong DING, Choubo SHEN, Chunhua HENGEL, Anton Van Den Existing anomaly detection paradigms overwhelmingly focus on training detection models using exclusively normal data or unlabeled data (mostly normal samples). One notorious issue with these approaches is that they are weak in discriminating anomalies from normal samples due to the lack of the knowledge about the anomalies. Here, we study the problem of few-shot anomaly detection, in which we aim at using a few labeled anomaly examples to train sample-efficient discriminative detection models. To address this problem, we introduce a novel weakly-supervised anomaly detection framework to train detection models without assuming the examples illustrating all possible classes of anomaly.Specifically, the proposed approach learns discriminative normality (regularity) by leveraging the labeled anomalies and a prior probability to enforce expressive representations of normality and unbounded deviated representations of abnormality. This is achieved by an end-to-end optimization of anomaly scores with a neural deviation learning, in which the anomaly scores of normal samples are imposed to approximate scalar scores drawn from the prior while that of anomaly examples is enforced to have statistically significant deviations from these sampled scores in the upper tail. Furthermore, our model is optimized to learn fine-grained normality and abnormality by top-K multiple-instance-learning-based feature subspace deviation learning, allowing more generalized representations. Comprehensive experiments on nine real-world image anomaly detection benchmarks show that our model is substantially more sample-efficient and robust, and performs significantly better than state-of-the-art competing methods in both closed-set and open-set settings. Our model can also offer explanation capability as a result of its prior-driven anomaly score learning. Code and datasets are available at: this https URL. 2021-08-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/7036 https://ink.library.smu.edu.sg/context/sis_research/article/8039/viewcontent/2108.00462.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Artificial Intelligence and Robotics OS and Networks
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Artificial Intelligence and Robotics
OS and Networks
spellingShingle Artificial Intelligence and Robotics
OS and Networks
PANG, Guansong
DING, Choubo
SHEN, Chunhua
HENGEL, Anton Van Den
Explainable deep few-shot anomaly detection with deviation networks
description Existing anomaly detection paradigms overwhelmingly focus on training detection models using exclusively normal data or unlabeled data (mostly normal samples). One notorious issue with these approaches is that they are weak in discriminating anomalies from normal samples due to the lack of the knowledge about the anomalies. Here, we study the problem of few-shot anomaly detection, in which we aim at using a few labeled anomaly examples to train sample-efficient discriminative detection models. To address this problem, we introduce a novel weakly-supervised anomaly detection framework to train detection models without assuming the examples illustrating all possible classes of anomaly.Specifically, the proposed approach learns discriminative normality (regularity) by leveraging the labeled anomalies and a prior probability to enforce expressive representations of normality and unbounded deviated representations of abnormality. This is achieved by an end-to-end optimization of anomaly scores with a neural deviation learning, in which the anomaly scores of normal samples are imposed to approximate scalar scores drawn from the prior while that of anomaly examples is enforced to have statistically significant deviations from these sampled scores in the upper tail. Furthermore, our model is optimized to learn fine-grained normality and abnormality by top-K multiple-instance-learning-based feature subspace deviation learning, allowing more generalized representations. Comprehensive experiments on nine real-world image anomaly detection benchmarks show that our model is substantially more sample-efficient and robust, and performs significantly better than state-of-the-art competing methods in both closed-set and open-set settings. Our model can also offer explanation capability as a result of its prior-driven anomaly score learning. Code and datasets are available at: this https URL.
format text
author PANG, Guansong
DING, Choubo
SHEN, Chunhua
HENGEL, Anton Van Den
author_facet PANG, Guansong
DING, Choubo
SHEN, Chunhua
HENGEL, Anton Van Den
author_sort PANG, Guansong
title Explainable deep few-shot anomaly detection with deviation networks
title_short Explainable deep few-shot anomaly detection with deviation networks
title_full Explainable deep few-shot anomaly detection with deviation networks
title_fullStr Explainable deep few-shot anomaly detection with deviation networks
title_full_unstemmed Explainable deep few-shot anomaly detection with deviation networks
title_sort explainable deep few-shot anomaly detection with deviation networks
publisher Institutional Knowledge at Singapore Management University
publishDate 2021
url https://ink.library.smu.edu.sg/sis_research/7036
https://ink.library.smu.edu.sg/context/sis_research/article/8039/viewcontent/2108.00462.pdf
_version_ 1770576192438534144