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

Gong, Daozhi; Liu, Lingqiao; Le, Vuong; Saha, Budhaditya; Mansour, Moussa Reda; Venkatesh, Svetha; Hengel, Anton van den

doi:10.48550/arxiv.1904.02639

Cited by 21 publications

(32 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Concretely, we encode the spatial information in multi-sensor time-series signals into the low-dimensional representation via Deep Convolutional Autoencoder (CAE). In order to reduce the effect of noisy data, some existing works have tried to add Memory module [24] or Gaussian Mixture Model (GMM) [23]. In our proposed method, we simplify these modules into penalty item, which called Maximum Mean Discrepancy (MMD) penalty.…”

Section: Overviewmentioning

confidence: 99%

“…It is generally assumed that the compression of anomalous samples is different from that on normal samples, and the reconstruction error becomes higher for these anomalous samples. In reality, being influenced by the high complexity of model and the noise of data, the reconstruction error for the abnormal input could also be fit so well by the training model [23], [24]. That is, the model is robust to noise and anomalies.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Unsupervised Deep Anomaly Detection for Multi-Sensor Time-Series Signals

Zhang¹,

Chen

Wang

et al. 2021

Preprint

View full text Add to dashboard Cite

Nowadays, multi-sensor technologies are applied in many fields, e.g., Health Care (HC), Human Activity Recognition (HAR), and Industrial Control System (ICS). These sensors can generate a substantial amount of multivariate time-series data. Unsupervised anomaly detection on multi-sensor time-series data has been proven critical in machine learning researches. The key challenge is to discover generalized normal patterns by capturing spatial-temporal correlation in multi-sensor data. Beyond this challenge, the noisy data is often intertwined with the training data, which is likely to mislead the model by making it hard to distinguish between the normal, abnormal, and noisy data. Few of previous researches can jointly address these two challenges. In this paper, we propose a novel deep learning-based anomaly detection algorithm called Deep Convolutional Autoencoding Memory network (CAE-M). We first build a Deep Convolutional Autoencoder to characterize spatial dependence of multi-sensor data with a Maximum Mean Discrepancy (MMD) to better distinguish between the noisy, normal, and abnormal data. Then, we construct a Memory Network consisting of linear (Autoregressive Model) and non-linear predictions (Bidirectional LSTM with Attention) to capture temporal dependence from time-series data. Finally, CAE-M jointly optimizes these two subnetworks. We empirically compare the proposed approach with several state-of-the-art anomaly detection methods on HAR and HC datasets. Experimental results demonstrate that our proposed model outperforms these existing methods.

show abstract

Section: Overviewmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Unsupervised Deep Anomaly Detection for Multi-Sensor Time-Series Signals

Zhang¹,

Chen

Wang

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Zero-shot anomaly detection methods assume that the conditional generative model [3,15,31,26,25,21,14]) can only reconstruct normal data. Hence, when presented with an abnormal test image, the model produces a large reconstruction error.…”

Section: Related Workmentioning

confidence: 99%

“…Even though they are often effective, these approaches still need a fairly high number of abnormal training images. Alternatively, zero-shot anomaly detection methods [3,15,31,14] Fig. 1.…”

Section: Introductionmentioning

confidence: 99%

“…In practice, FSAD-NET needs significantly less abnormal training images than typical imbalanced learning problems [10,11]. Moreover, given that we access a few abnormal training images, FSAD-NET has the potential to be more effective at correctly classifying challenging outliers compared to typical zeroshot anomaly detection methods [3,15,31,14]. To the best of our knowledge, our method is the first medical image analysis work to explore few-shot anomaly detection with a feature encoder that maximises MI between training images and embeddings, and explicitly optimises anomaly scores.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Few-Shot Anomaly Detection for Polyp Frames from Colonoscopy

Tian

Maicas

et al. 2020

Preprint

View full text Add to dashboard Cite

5 Anomaly detection methods generally target the learning of a normal image distribution (i.e., inliers showing healthy cases) and during testing, samples relatively far from the learned distribution are classified as anomalies (i.e., outliers showing disease cases). These approaches tend to be sensitive to outliers that lie relatively close to inliers (e.g., a colonoscopy image with a small polyp). In this paper, we address the inappropriate sensitivity to outliers by also learning from inliers. We propose a new few-shot anomaly detection method based on an encoder trained to maximise the mutual information between feature embeddings and normal images, followed by a few-shot score inference network, trained with a large set of inliers and a substantially smaller set of outliers. We evaluate our proposed method on the clinical problem of detecting frames containing polyps from colonoscopy video sequences, where the training set has 13350 normal images (i.e., without polyps) and less than 100 abnormal images (i.e., with polyps). The results of our proposed model on this data set reveal a state-of-the-art detection result, while the performance based on different number of anomaly samples is relatively stable after approximately 40 abnormal training images. Code is available at https://github.com/tianyu0207/FSAD-Net.

show abstract

Software system to predict the infection in COVID‐19 patients using deep learning and web of things

et al. 2021

View full text Add to dashboard Cite

Since the end of 2019, computed tomography (CT) images have been used as an important substitute for the time‐consuming Reverse Transcriptase polymerase chain reaction (RT‐PCR) test; a new coronavirus 2019 (COVID‐19) disease has been detected and has quickly spread through many countries across the world. Medical imaging such as computed tomography provides great potential due to growing skepticism toward the sensitivity of RT‐PCR as a screening tool. For this purpose, automated image segmentation is highly desired for a clinical decision aid and disease monitoring. However, there is limited publicly accessible COVID‐19 image knowledge, leading to the overfitting of conventional approaches. To address this issue, the present paper focuses on data augmentation techniques to create synthetic data. Further, a framework has been proposed using WoT and traditional U‐Net with EfficientNet B0 to segment the COVID Radiopedia and Medseg datasets automatically. The framework achieves an F ‐score of 0.96, which is best among state‐of‐the‐art methods. The performance of the proposed framework also computed using Sensitivity, Specificity, and Dice‐coefficient, achieves 84.5%, 93.9%, and 65.0%, respectively. Finally, the proposed work is validated using three quality of service (QoS) parameters such as server latency, response time, and network latency which improves the performance by 8%, 7%, and 10%, respectively.

show abstract

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

Cited by 21 publications

References 29 publications

Unsupervised Deep Anomaly Detection for Multi-Sensor Time-Series Signals

Unsupervised Deep Anomaly Detection for Multi-Sensor Time-Series Signals

Few-Shot Anomaly Detection for Polyp Frames from Colonoscopy

Software system to predict the infection in COVID‐19 patients using deep learning and web of things

Contact Info

Product

Resources

About