373 Hits in 7.1 sec

Deep Learning for Anomaly Detection: A Survey [article]

Raghavendra Chalapathy (University of Sydney and Capital Markets Cooperative Research Centre, Sanjay Chawla (Qatar Computing Research Institute
2019 arXiv   pre-print
Within each category we outline the basic anomaly detection technique, along with its variants and present key assumptions, to differentiate between normal and anomalous behavior.  ...  The aim of this survey is two-fold, firstly we present a structured and comprehensive overview of research methods in deep learning-based anomaly detection.  ...  Multi-variate time series deep anomaly detection Anomaly detection in multivariate time series data is a challenging task.  ... 
arXiv:1901.03407v2 fatcat:x3tb4ccxfvdkfo7k2y2oxhr7ly

Recent Advances in Anomaly Detection Methods Applied to Aviation

Luis Basora, Xavier Olive, Thomas Dubot
2019 Aerospace (Basel)  
In particular, we cover unsupervised techniques applicable to time series data because of their relevance to the aviation domain, where the lack of labeled data is the most usual case, and the nature of  ...  Anomaly detection is an active area of research with numerous methods and applications.  ...  Recurrent Neural Networks autoencoders have also been used on time series [125] in order to find a proper embedding or representation of time series that is in turn used for predicting a RUL estimation  ... 
doi:10.3390/aerospace6110117 fatcat:kprkb643xrhcnmjy2c2lbzoa7m

Unsupervised Anomaly Detection of Industrial Robots using Sliding-Window Convolutional Variational Autoencoder

Tingting Chen, Xueping Liu, Bizhong Xia, Wei Wang, Yongzhi Lai
2020 IEEE Access  
In this paper, we introduce an unsupervised anomaly detection for industrial robots, sliding-window convolutional variational autoencoder (SWCVAE), which can realize real-time anomaly detection spatially  ...  and temporally by coping with multivariate time series data.  ...  Pereira and Silveira [27] proposed an unsupervised anomaly detection method using variational recurrent autoencoders with attention which was applied to energy time series data.  ... 
doi:10.1109/access.2020.2977892 fatcat:5a74eirhvfdkdho2mb7uzz7ehy

Digital Twins in Solar Farms: An Approach through Time Series and Deep Learning

Kamel Arafet, Rafael Berlanga
2021 Algorithms  
The resulting data are used to train a DL model (e.g., autoencoders) in order to detect anomalies of the physical system in its DT.  ...  To build such a DT, sensor-based time series are properly analyzed and processed.  ...  Several studies have shown the efficacy of both in the detection of anomalies [10, 11] . Another DL family widely used for time series is the autoencoders.  ... 
doi:10.3390/a14050156 fatcat:5s5nqzwwora2jnfgwkd2yytvgm

Towards delicate anomaly detection of energy consumption for buildings: enhance the performance from two levels

Dong Wang, Therese Enlund, Johan Trygg, Mats Tysklind, Lili Jiang
2022 IEEE Access  
As the most flexible and applicable type of anomaly detection approach, unsupervised anomaly detection has been implemented in several studies for building energy data.  ...  First, it precisely detected the contextual anomalies concealed beneath the time variation of the energy consumption profiles of the three buildings.  ...  Pereira et al. proposed an approach incorporating variational recurrent autoencoder with self-attention and probabilistic reconstruction scoring for anomaly detection of time series energy consumption  ... 
doi:10.1109/access.2022.3160170 fatcat:fkmdmqrlhze6ngf5r2ej7ho5oi

Artificial Intelligence based Anomaly Detection of Energy Consumption in Buildings: A Review, Current Trends and New Perspectives [article]

Yassine Himeur and Khalida Ghanem and Abdullah Alsalemi and Faycal Bensaali and Abbes Amira
2020 arXiv   pre-print
If leveraged properly, that data could assist end-users, energy producers and utility companies in detecting anomalous power consumption and understanding the causes of each anomaly.  ...  In this regard, this paper an in-depth review of existing anomaly detection frameworks for building energy consumption based on artificial intelligence.  ...  Similarly, in [61] , the authors detect anomalies in time-series power footprints using variational recurrent autoencoder.  ... 
arXiv:2010.04560v4 fatcat:dpullqvuv5f5lhu6tyqgdbya3q

Deep Learning Based Anomaly Detection for Muti-dimensional Time Series: A Survey [chapter]

Zhipeng Chen, Zhang Peng, Xueqiang Zou, Haoqi Sun
2022 Communications in Computer and Information Science  
It also has the ability to describe space and time and is widely used in many fields such as system state anomaly detection.  ...  In the big data scenario, deep learning method begins to be applied to anomaly detection tasks for multi-dimensional time series due to its wide coverage and strong learning ability.  ...  In recent years, some work tends to use semi supervised or unsupervised methods to detect anomalies in time series.  ... 
doi:10.1007/978-981-16-9229-1_5 fatcat:te5p2vz2hfcu7cydy3vrq7fkye

Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detection [article]

Dong Gong, Lingqiao Liu, Vuong Le, Budhaditya Saha, Moussa Reda Mansour, Svetha Venkatesh, Anton van den Hengel
2019 arXiv   pre-print
Deep autoencoder has been extensively used for anomaly detection.  ...  To mitigate this drawback for autoencoder based anomaly detector, we propose to augment the autoencoder with a memory module and develop an improved autoencoder called memory-augmented autoencoder, i.e  ...  Related Work Anomaly detection In unsupervised anomaly detection, only normal samples are available as training data [3] .  ... 
arXiv:1904.02639v2 fatcat:bki7ibp3fnccljokd7erhkyk44

Deep Learning for Anomaly Detection in Time-Series Data: Review, Analysis, and Guidelines

Kukjin Choi, Jihun Yi, Changhwa Park, Sungroh Yoon
2021 IEEE Access  
This review provides a background on anomaly detection in time-series data and reviews the latest applications in the real world.  ...  Also, we comparatively analyze state-of-the-art deep-anomaly-detection models for time series with several benchmark datasets.  ...  If the time-series data has a seasonal-or cyclic-variation, we can use a seasonal ARIMA (SARIMA) [81] model.  ... 
doi:10.1109/access.2021.3107975 fatcat:yrlegcnsy5d47ds3vgbzq64qcu

Deep Neural Network Classifier for Variable Stars with Novelty Detection Capability [article]

Benny T.-H. Tsang, William C. Schultz
2019 arXiv   pre-print
We present a periodic light curve classifier that combines a recurrent neural network autoencoder for unsupervised feature extraction and a dual-purpose estimation network for supervised classification  ...  The simultaneous training of the autoencoder and estimation network is found to be mutually beneficial, resulting in faster autoencoder convergence, and superior classification and novelty detection performance  ...  ACKNOWLEDGEMENTS This research project has benefited from interactions with Tom Prince, Thomas Kupfer, and Jan van Roestel.  ... 
arXiv:1905.05767v1 fatcat:jtnpo65ihnb7dbnrfl76fzqzxa

Self-Adaption AAE-GAN for Aluminum Electrolytic Cell Anomaly Detection

Danyang Cao, Di Liu, Xu Ren, Nan Ma
2021 IEEE Access  
This time series anomaly detection method converts multi-dimensional time series data into a two-dimensional matrix, and only normal samples are needed in the training process, which effectively solves  ...  INDEX TERMS Aluminum electrolytic cell, anomaly detection, AAE-GAN, multivariate time series, imbalanced industrial time series.  ...  After this method was proposed, technologies such as sliding window and variational autoencoder (VAE) were also applied to this idea for anomaly detection of time series data.  ... 
doi:10.1109/access.2021.3097116 fatcat:ufxi7wohc5hjjd7esyv3uksd5q

Latent Space Autoregression for Novelty Detection

Davide Abati, Angelo Porrello, Simone Calderara, Rita Cucchiara
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
one-class and video anomaly detection settings.  ...  In our proposal, we design a general framework where we equip a deep autoencoder with a parametric density estimator that learns the probability distribution underlying its latent representations through  ...  We gratefully acknowledge Facebook Artificial Intelligence Research and Panasonic Silicon Valley Lab for the donation of GPUs used for this research.  ... 
doi:10.1109/cvpr.2019.00057 dblp:conf/cvpr/AbatiPCC19 fatcat:tvlifcymbfcfvaqfq3l4fanmti

Energy-based Models for Video Anomaly Detection [article]

Hung Vu, Dinh Phung, Tu Dinh Nguyen, Anthony Trevors, Svetha Venkatesh
2017 arXiv   pre-print
This allows our system to be trained completely in an unsupervised procedure and liberate us from the need for costly data annotation.  ...  Instead of hanlding with ambiguous definition of anomaly objects, we propose to work with regular patterns whose unlabeled data is abundant and usually easy to collect in practice.  ...  The idea of variational autoencoders was extended to sequential data in variational RNN (Chung et al., 2015) and discrete data in discrete variational autoencoders (Rolfe, 2016) .  ... 
arXiv:1708.05211v1 fatcat:fkb2m2vdx5fs5grz3mbbmlyctu

A Review of Deep Learning Algorithms and Their Applications in Healthcare

Hussein Abdel-Jaber, Disha Devassy, Azhar Al Salam, Lamya Hidaytallah, Malak EL-Amir
2022 Algorithms  
As a review, this paper aims to categorically cover several widely used deep learning algorithms along with their architectures and their practical applications: backpropagation, autoencoders, variational  ...  It was found that deep learning approaches can be used for big data analysis successfully.  ...  Anomaly detection Data compression • Feature variation • For learning latent representations Variational Autoen-coders Unsupervised • A better organization of the latent space results in high quality  ... 
doi:10.3390/a15020071 fatcat:ku5mfuijdjfxxdv7hlkexad7dy

Deep anomaly detection for industrial systems: a case study

Feng Xue, Weizhong Yan, Tianyi Wang, Hao Huang, Bojun Feng
2020 Proceedings of the Annual Conference of the Prognostics and Health Management Society, PHM  
We explore the use of deep neural networks for anomaly detection of industrial systems where the data are multivariate time series measurements.  ...  Also, Support Vector Data Description (SVDD) method is adapted to such anomaly detection settings with deep neural networks.  ...  ACKNOWLEDGMENT This material is based upon work supported by the Department of Energy, National Energy Technology Laboratory under Award Number DE-FE0031763.  ... 
doi:10.36001/phmconf.2020.v12i1.1186 fatcat:mtseixwucjebdjbnzli356c6rm
« Previous Showing results 1 — 15 out of 373 results