Robust Anomaly Detection in Time Series through Variational AutoEncoders and a Local Similarity Score

Matias, Pedro; Folgado, Duarte; Gamboa, Hugo; Carreiro, André V.

doi:10.5220/0010320500910102

Cited by 11 publications

(5 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We therefore group all the 4 abnormal classes into one class. Thus, the ECG5000 timeseries classification task is modeled as a binary classification task between healthy and unhealthy heartbeats; a few authors do the same (Matias et al, 2021;Oluwasanmi et al, 2021;Biloborodova et al, 2022).…”

Section: Ecgmentioning

confidence: 99%

Reservoir based spiking models for univariate Time Series Classification

Gaurav

Stewart

2023

Front. Comput. Neurosci.

View full text Add to dashboard Cite

A variety of advanced machine learning and deep learning algorithms achieve state-of-the-art performance on various temporal processing tasks. However, these methods are heavily energy inefficient—they run mainly on the power hungry CPUs and GPUs. Computing with Spiking Networks, on the other hand, has shown to be energy efficient on specialized neuromorphic hardware, e.g., Loihi, TrueNorth, SpiNNaker, etc. In this work, we present two architectures of spiking models, inspired from the theory of Reservoir Computing and Legendre Memory Units, for the Time Series Classification (TSC) task. Our first spiking architecture is closer to the general Reservoir Computing architecture and we successfully deploy it on Loihi; the second spiking architecture differs from the first by the inclusion of non-linearity in the readout layer. Our second model (trained with Surrogate Gradient Descent method) shows that non-linear decoding of the linearly extracted temporal features through spiking neurons not only achieves promising results, but also offers low computation-overhead by significantly reducing the number of neurons compared to the popular LSM based models—more than 40x reduction with respect to the recent spiking model we compare with. We experiment on five TSC datasets and achieve new SoTA spiking results (—as much as 28.607% accuracy improvement on one of the datasets), thereby showing the potential of our models to address the TSC tasks in a green energy-efficient manner. In addition, we also do energy profiling and comparison on Loihi and CPU to support our claims.

show abstract

Section: Ecgmentioning

confidence: 99%

Reservoir based spiking models for univariate Time Series Classification

Gaurav

Stewart

2023

Front. Comput. Neurosci.

View full text Add to dashboard Cite

show abstract

“…Therefore, we decided to opt for an anomaly detection task that requires fewer data for training. Interestingly, Variational Autoencoders have been successfully employed for anomaly/change/novelty detection tasks, such as bird species [19], ultrasounds [16], time series [14], computer networks [11], etc.…”

Section: # Of Observationsmentioning

confidence: 99%

Variational Autoencoders for Anomaly Detection in Respiratory Sounds

Cozzatti¹,

Simonetta²,

Ntalampiras³

2022

Preprint

View full text Add to dashboard Cite

This paper proposes a weakly-supervised machine learningbased approach aiming at a tool to alert patients about possible respiratory diseases. Various types of pathologies may affect the respiratory system, potentially leading to severe diseases and, in certain cases, death. In general, effective prevention practices are considered as major actors towards the improvement of the patient's health condition. The proposed method strives to realize an easily accessible tool for the automatic diagnosis of respiratory diseases. Specifically, the method leverages Variational Autoencoder architectures permitting the usage of training pipelines of limited complexity and relatively small-sized datasets. Importantly, it offers an accuracy of 57%, which is in line with the existing strongly-supervised approaches.

show abstract

“…More recently, DL models have started to be applied in time series analysis (concretely concerning classification and anomaly detection tasks) [26][27][28][29][30] after having achieved a notable success in the computer vision field [31]. DNN models offer many advantages when compared to classical approaches, since they do not require an elaborate data pre-processing pipeline, they are capable of efficiently extract relevant feature maps (unlike hand-crafted methods, which require expert domain knowledge and can be computationally more expensive to extract [32,33]), and better handle abundant amounts of data.…”

Section: Related Workmentioning

confidence: 99%

Time Series Segmentation Using Neural Networks with Cross-Domain Transfer Learning

et al. 2021

Self Cite

View full text Add to dashboard Cite

Searching for characteristic patterns in time series is a topic addressed for decades by the research community. Conventional subsequence matching techniques usually rely on the definition of a target template pattern and a searching method for detecting similar patterns. However, the intrinsic variability of time series introduces changes in patterns, either morphologically and temporally, making such techniques not as accurate as desired. Intending to improve segmentation performances, in this paper, we proposed a Mask-based Neural Network (NN) which is capable of extracting desired patterns of interest from long time series, without using any predefined template. The proposed NN has been validated, alongside a subsequence matching algorithm, in two datasets: clinical (electrocardiogram) and human activity (inertial sensors). Moreover, the reduced dimension of the data in the latter dataset led to the application of transfer learning and data augmentation techniques to reach model convergence. The results have shown the proposed model achieved better segmentation performances than the baseline one, in both domains, reaching average Precision and Recall scores of 99.0% and 97.5% (clinical domain), along with 77.0% and 71.4% (human activity domain), introducing Neural Networks and Transfer Learning as promising alternatives for pattern searching in time series.

show abstract

Robust Anomaly Detection in Time Series through Variational AutoEncoders and a Local Similarity Score

Cited by 11 publications

References 11 publications

Reservoir based spiking models for univariate Time Series Classification

Reservoir based spiking models for univariate Time Series Classification

Variational Autoencoders for Anomaly Detection in Respiratory Sounds

Time Series Segmentation Using Neural Networks with Cross-Domain Transfer Learning

Contact Info

Product

Resources

About