A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf
.
Filters
Deep Contrastive One-Class Time Series Anomaly Detection
[article]
2022
arXiv
pre-print
To overcome the shortcomings, a deep Contrastive One-Class Anomaly detection method of time series (COCA) is proposed by authors, following the normality assumptions of CL and one-class classification. ...
The accumulation of time-series data and the absence of labels make time-series Anomaly Detection (AD) a self-supervised deep learning task. ...
and one-class classification-based. ...
arXiv:2207.01472v2
fatcat:x2kb6ypbs5asddjvvtpsxyxdri
Neural Transformation Learning for Deep Anomaly Detection Beyond Images
[article]
2022
arXiv
pre-print
Extensive experiments on time series demonstrate that our proposed method outperforms existing approaches in the one-vs.-rest setting and is competitive in the more challenging n-vs. ...
On tabular datasets from the medical and cyber-security domains, our method learns domain-specific transformations and detects anomalies more accurately than previous work. ...
Neural Transformation Learning for Deep Anomaly Detection We develop neural transformation learning for anomaly detection (NeuTraL AD), a deep anomaly detection method based on contrastive learning for ...
arXiv:2103.16440v4
fatcat:kr6st3ulyfgvfpcpdkr5s4esra
TS-Rep: Time series representation learning in robotics using self-supervised contrastive learning
2022
Zenodo
We further train a one-class classifier on output representation for anomaly detection task and achieve significant improvement over state-of-the-art anomaly detection frameworks. ...
We propose a new self-supervised contrastive learning framework, TS-Rep, for time series representation learning in robotics. ...
Clusterability Anomaly Detection • Triplet loss • Time-based negative sampling and, • Positive sampling using nearest ...
doi:10.5281/zenodo.7135205
fatcat:n7uw2ao72fgdzjfef2ia64hm3m
DACAD: Domain Adaptation Contrastive Learning for Anomaly Detection in Multivariate Time Series
[article]
2024
arXiv
pre-print
In this paper, we propose a novel Domain Adaptation Contrastive learning for Anomaly Detection in multivariate time series (DACAD) model to address this issue by combining UDA and contrastive representation ...
Time series anomaly detection (TAD) faces a significant challenge due to the scarcity of labelled data, which hinders the development of accurate detection models. ...
In our study, we introduce Domain Adaptation Contrastive learning model for Anomaly Detection in time series (DA-CAD), a unique framework for UDA in multivariate time series anomaly detection, leveraging ...
arXiv:2404.11269v1
fatcat:ga2awran7nablptx553a44feeq
Calibrated One-class Classification for Unsupervised Time Series Anomaly Detection
[article]
2024
arXiv
pre-print
To tackle these problems, this paper proposes calibrated one-class classification for anomaly detection, realizing contamination-tolerant, anomaly-informed learning of data normality via uncertainty modeling-based ...
Time series anomaly detection is instrumental in maintaining system availability in various domains. ...
CONCLUSIONS This paper introduces COUTA, an unsupervised time series anomaly detection method based on calibrated one-class classification. ...
arXiv:2207.12201v2
fatcat:zezxq76kifbjhlv7unz7cik54q
Time series anomaly detection based on shapelet learning
2018
Computational statistics (Zeitschrift)
This article presents a novel method for unsupervised anomaly detection based on the shapelet transformation for time series. ...
Our approach learns representative features that describe the shape of time series stemming from the normal class, and simultaneously learns to accurately detect anomalous time series. ...
The remainder of this article is organized as follows: Sect. 2 starts with an overview of related work on time series anomaly detection and shapelet-based methods for time series classification. ...
doi:10.1007/s00180-018-0824-9
fatcat:x6tfzrqvtfblpnuf7thpzms3by
Active Learning for LSTM-autoencoder-based Anomaly Detection in Electrocardiogram Readings
2020
European Conference on Principles of Data Mining and Knowledge Discovery
Active learning for LSTM-autonecoder-based anomaly detection LSTM-autoencoder [9] is nowadays increasingly used to detect anomalies in time series data [11, 5, 4] . ...
Conclusion We presented an active Learning for LSTM-autoencoder-based anomaly detection for time-series data. ...
dblp:conf/pkdd/SabataH20
fatcat:hl7rywc4abe3zi4ndem7o6iwme
Learning Representation for Anomaly Detection of Vehicle Trajectories
[article]
2023
arXiv
pre-print
Different from general time-series anomaly detection, anomalous vehicle trajectory detection deals with much richer contexts on the road and fewer observable patterns on the anomalous trajectories themselves ...
We conduct extensive experiments to demonstrate that our supervised method based on contrastive learning and unsupervised method based on reconstruction with semantic latent space can significantly improve ...
Most previous works directly use VAE-based reconstruction, one-class SVM, or contrastive learning (unsupervised) to detect anomalies for time-series data. ...
arXiv:2303.05000v1
fatcat:uij5qp3pzzgbjmjotvetggw2ha
Research and application of Transformer based anomaly detection model: A literature review
[article]
2024
arXiv
pre-print
To inspire research on Transformer-based anomaly detection, this review offers a fresh perspective on the concept of anomaly detection. ...
The review includes an extensive compilation of over 100 core references related to Transformer-based anomaly detection. ...
Acknowledgement The authors thank the anonymous reviewers for their insightful suggestions on this work. ...
arXiv:2402.08975v1
fatcat:hxwevh2hivakfjpqaibwlzobby
Soft Contrastive Learning for Time Series
[article]
2024
arXiv
pre-print
In experiments, we demonstrate that SoftCLT consistently improves the performance in various downstream tasks including classification, semi-supervised learning, transfer learning, and anomaly detection ...
Contrastive learning has shown to be effective to learn representations from time series in a self-supervised way. ...
of false positives. • DONUT (Xu et al., 2018) : DONUT is an unsupervised anomaly detection algorithm based on variational autoencoder. • SR (Ren et al., 2019) : SR is a time-series anomaly detection ...
arXiv:2312.16424v3
fatcat:laq3nhyeijbxbilgcyhzuby3pq
Time Series Data Augmentation for Deep Learning: A Survey
[article]
2021
arXiv
pre-print
However, the labeled data of many real-world time series applications may be limited such as classification in medical time series and anomaly detection in AIOps. ...
Deep learning performs remarkably well on many time series analysis tasks recently. ...
Preliminary Evaluation
Time Series Anomaly Detection Given the challenges of both data scarcity and data imbalance in time series anomaly detection, it is crucial to make use of time series data augmentation ...
arXiv:2002.12478v2
fatcat:rhyj67nrgje5pg66wmxg5qhyge
Multi-Task Self-Supervised Time-Series Representation Learning
[article]
2023
arXiv
pre-print
We evaluate the proposed framework on three downstream tasks: time-series classification, forecasting, and anomaly detection. ...
This strategy can encourage varied consistency of time-series representations depending on the positive pair selection and contrastive loss. ...
We conducted downstream evaluations on time-series classification, forecasting, and anomaly detection. ...
arXiv:2303.01034v1
fatcat:5kccfudaibaihfehl2cghxosqe
Intrinsic Anomaly Detection for Multi-Variate Time Series
[article]
2022
arXiv
pre-print
We introduce a novel, practically relevant variation of the anomaly detection problem in multi-variate time series: intrinsic anomaly detection. ...
Intrinsic anomalies are changes in the functional dependency structure between time series that represent an environment and time series that represent the internal state of a system that is placed in ...
closely related to our approach; (ii) approaches based on classical time series embeddings; and (iii) standard anomaly detection methods on time series. ...
arXiv:2206.14342v1
fatcat:ux576xzdzngfvkvemc7fxqlvmm
Anomaly Detection of Wind Turbine Time Series using Variational Recurrent Autoencoders
[article]
2021
arXiv
pre-print
We have evaluated our approach on a custom wind turbine time series dataset, for the two-classes problem (one normal versus one abnormal class), we obtained a classification accuracy of up to 96% on test ...
In this work, we investigate the problem of ice accumulation in wind turbines by framing it as anomaly detection of multi-variate time series. ...
Anomaly detection We perform anomaly scoring using the learned lowdimensional time series representations provided by the VRAE model. ...
arXiv:2112.02468v1
fatcat:bp4lf2s6lbcg3leltrn3dz7swm
Detection of anomalous ticket purchasing behavior for concerts based on machine learning
2024
Applied and Computational Engineering
an overall anomaly detection system. ...
This paper focuses on the use of automated systems, robots, or malicious software for ticket purchases, limiting consumers participation in abnormal ticket-buying activities. ...
paper "Unsupervised anomaly detection by densely contrastive learning for time series data." ...
doi:10.54254/2755-2721/51/20241495
fatcat:envby6ha2rf2pnx2a56l3wva6e
« Previous
Showing results 1 — 15 out of 24,087 results