Filters








4,333 Hits in 3.6 sec

Context-encoding Variational Autoencoder for Unsupervised Anomaly Detection – Short Paper [article]

David Zimmerer, Simon Kohl, Jens Petersen, Fabian Isensee, Klaus Maier-Hein
2020 arXiv   pre-print
In this context, deep learning-based autoencoders have shown great potential in detecting anomalies in medical images.  ...  Unsupervised learning can leverage large-scale data sources without the need for annotations.  ...  Our results suggest that these features can improve out-of-distribution/anomaly-detection tasks and as such aid VAEs in capturing the data distribution.  ... 
arXiv:1907.12258v2 fatcat:xoxvcp6tb5ehbhaojoj4s635bm

Outlier Detection in High Dimensional Data

Anusha L., Nagaraja G S
2021 International Journal of Engineering and Advanced Technology  
The paper proposed on outlier detection for multivariate high dimensional data for Autoencoder unsupervised model.  ...  Machine Learning is a branch of Artificial Intelligence that allows a machine to learn and improve at a task over time.  ...  H) Detect anomalies for the datasets: Based on the threshold value the anomaly has been detected.  ... 
doi:10.35940/ijeat.e2675.0610521 fatcat:pv4mautonzfwlkd5p4khmh7qh4

Anomaly Detection in Industrial Software Systems - Using Variational Autoencoders

Tharindu Kumarage, Nadun De Silva, Malsha Ranawaka, Chamal Kuruppu, Surangika Ranathunga
2018 Proceedings of the 7th International Conference on Pattern Recognition Applications and Methods  
Nevertheless, the high complexity of the system components makes anomaly detection a high dimensional machine learning problem.  ...  We show that, when used in an unsupervised manner, VAE outperforms the well-known clustering technique DBSCAN.  ...  ACKNOWLEDGEMENTS The authors thank MillenniumIT Software (Private) Ltd. for providing the datasets for industrial system evaluation.  ... 
doi:10.5220/0006600304400447 dblp:conf/icpram/KumarageSRKR18 fatcat:kvfm6f2uajd7nhklyglvgjrvpa

Anomalous Sound Detection using unsupervised and semi-supervised autoencoders and gammatone audio representation [article]

Sergi Perez-Castanos, Javier Naranjo-Alcazar, Pedro Zuccarello, Maximo Cobos
2020 arXiv   pre-print
For example, related to industrial processes, the early detection of malfunctions or damage in machines can mean great savings and an improvement in the efficiency of industrial processes.  ...  Unsupervised detection is attracting a lot of interest due to its immediate applicability in many fields.  ...  However, a more intersting and complex approach was chosen for this work: one single anomaly detector (in this case autoencoder) was trained for all the machines.  ... 
arXiv:2006.15321v1 fatcat:xmm4few4lbcebdahaoquee7bkm

Unsupervised log message anomaly detection

Amir Farzad, T. Aaron Gulliver
2020 ICT Express  
In this paper, an unsupervised model for log message anomaly detection is proposed which employs Isolation Forest and two deep Autoencoder networks.  ...  The Autoencoder networks are used for training and feature extraction, and then for anomaly detection, while Isolation Forest is used for positive sample prediction.  ...  In this paper, a model was proposed for unsupervised anomaly detection using Isolation Forest and two deep Autoencoder networks. These networks are used for feature extraction and anomaly detection.  ... 
doi:10.1016/j.icte.2020.06.003 fatcat:qs6xnknhi5gajad23gtlmwrqxq

Using Autoencoders for Anomaly Detection and Transfer Learning in IoT

Chin-Wei Tien, Tse-Yung Huang, Ping-Chun Chen, Jenq-Haur Wang
2021 Computers  
Finally, we evaluate the performance of anomaly detection by transfer learning with autoencoders.  ...  Comparable performance of anomaly detection can be achieved when using autoencoders for transfer learning from the reference dataset in the literature to our target site.  ...  Acknowledgments: This study is conducted under the "Artificial Intelligence Oriented for Cyber Security Technology Collaboration Project (1/4)" of the Institute of Information Industry which is subsidized  ... 
doi:10.3390/computers10070088 fatcat:acd2tgu2ubf7pdyvjdrfss6kj4

Active Learning for LSTM-autoencoder-based Anomaly Detection in Electrocardiogram Readings

Tomás Sabata, Martin Holena
2020 European Conference on Principles of Data Mining and Knowledge Discovery  
Figure 1 shows how actively asking for annotations can improve the unsupervised anomaly detection with an LSTM-autoencoder (red dashed line).  ...  For this reason, unsupervised machine learning techniques such as anomaly detection are often used with such data.  ... 
dblp:conf/pkdd/SabataH20 fatcat:hl7rywc4abe3zi4ndem7o6iwme

Unsupervised Anomaly Detection Based on Deep Autoencoding and Clustering

Chuanlei Zhang, Jiangtao Liu, Wei Chen, Jinyuan Shi, Minda Yao, Xiaoning Yan, Nenghua Xu, Dufeng Chen, Chi-Hua Chen
2021 Security and Communication Networks  
In order to improve the performance of unsupervised anomaly detection, we propose an anomaly detection scheme based on a deep autoencoder (DAE) and clustering methods.  ...  The key to anomaly detection is density estimation.  ...  Unsupervised Anomaly Detection Scheme.  ... 
doi:10.1155/2021/7389943 fatcat:zph3oio3izh3nbekzcjyyu6v2i

Visual Anomaly Detection for Images: A Survey [article]

Jie Yang, Ruijie Xu, Zhiquan Qi, Yong Shi
2021 arXiv   pre-print
In this paper, we provide a comprehensive survey of the classical and deep learning-based approaches for visual anomaly detection in the literature.  ...  Visual anomaly detection is an important and challenging problem in the field of machine learning and computer vision.  ...  [26] first use the deep autoencoder for anomaly detection of high-dimensional data.  ... 
arXiv:2109.13157v1 fatcat:zzut43xetng7tg3cghz25sjzoq

Unsupervised anomaly detection for a Smart Autonomous Robotic Assistant Surgeon (SARAS)using a deep residual autoencoder [article]

Dinesh Jackson Samuel, Fabio Cuzzolin
2021 arXiv   pre-print
In this work we thus propose an unsupervised approach to anomaly detection for robotic-assisted surgery based on deep residual autoencoders.  ...  The end-to-end system was developed and deployed as part of the SARAS demonstration platform for real-time anomaly detection with a processing time of about 25 ms per frame.  ...  METHODOLOGY The proposed methodology for anomaly detection is thus based on an unsupervised learning approach using a deep residual autoencoder.  ... 
arXiv:2104.11008v1 fatcat:bvpmpb7gjjfzzg35dixagf7b2m

Unsupervised Anomaly Detection for a Smart Autonomous Robotic Assistant Surgeon (SARAS) Using a Deep Residual Autoencoder

Dinesh Jackson Samuel, Fabio Cuzzolin
2021 IEEE Robotics and Automation Letters  
In this work we propose an unsupervised approach to anomaly detection for robotic MIS based on deep residual autoencoders.  ...  The system was developed and deployed as part of the SARAS platform for real-time anomaly detection with a processing time of 25 ms per frame.  ...  METHODOLOGY The proposed methodology for anomaly detection is thus based on an unsupervised learning approach using a deep residual autoencoder.  ... 
doi:10.1109/lra.2021.3097244 fatcat:2eiudxl4kbe7fgrbmwdgkys7ca

Transfer Learning from an Auxiliary Discriminative Task for Unsupervised Anomaly Detection [article]

Urwa Muaz, Stanislav Sobolevsky
2019 arXiv   pre-print
We propose use of an auxiliary classification task to extract features from unlabelled data by supervised learning, which can be used for unsupervised anomaly detection.  ...  Unsupervised anomaly detection from high dimensional data like mobility networks is a challenging task.  ...  Zong et al. (2018) uses a coupled pipeline of deep autoencoder and Gaussian Mixture Models for unsupervised anomaly detection.  ... 
arXiv:1912.02864v1 fatcat:kuzbwcfcfbaqbdlkjpnb233g3q

Diminishing Empirical Risk Minimization for Unsupervised Anomaly Detection [article]

Shaoshen Wang
2022 arXiv   pre-print
Unsupervised anomaly detection (AD) is a challenging task in realistic applications. Recently, there is an increasing trend to detect anomalies with deep neural networks (DNN).  ...  In this work, we identify one reason that hinders most existing DNN-based anomaly detection methods from performing is the wide adoption of the Empirical Risk Minimization (ERM).  ...  What's more, MemAE [11] proposes a memory-augmented autoencoder to improve the performance of unsupervised anomaly detection.  ... 
arXiv:2205.14676v1 fatcat:sbehbd4jdzctrhyw4i34avgrg4

Online-compatible Unsupervised Non-resonant Anomaly Detection [article]

Vinicius Mikuni, Benjamin Nachman, David Shih
2021 arXiv   pre-print
We propose the first complete strategy for unsupervised detection of non-resonant anomalies that includes both signal sensitivity and a data-driven method for background estimation.  ...  This method can be deployed offline for non-resonant anomaly detection and is also the first complete online-compatible anomaly detection strategy.  ...  ACKNOWLEDGMENTS We thank Barry Dillon and Gregor Kasieczka for feedback on the manuscript. VM and BN are supported by the U.S.  ... 
arXiv:2111.06417v1 fatcat:isjhh7dxbzf6na6xjsca3td7vi

A Comparison of Supervised and Unsupervised Deep Learning Methods for Anomaly Detection in Images [article]

Vincent Wilmet, Sauraj Verma, Tabea Redl, Håkon Sandaker, Zhenning Li
2021 arXiv   pre-print
We utilize the MVTec anomaly dataset and develop three different models, a CNN for supervised anomaly detection, KD-CAE for autoencoder anomaly detection, NI-CAE for noise induced anomaly detection and  ...  Therefore, in this paper we investigate different methods of deep learning, including supervised and unsupervised learning, for anomaly detection applied to a quality assurance use case.  ...  [11] developed a deep autoencoding gaussian mixture model (DAGMM) for unsupervised anomaly detection which utilizes a deep autoencoder alongside a gaussian mixture model, and found that their model  ... 
arXiv:2107.09204v1 fatcat:7dvntkgoubcy5kwrwvqksumswq
« Previous Showing results 1 — 15 out of 4,333 results