Anomaly Scores Research Articles

Energy Consumption Auditing Anomaly Detection System of Robotic Manipulators based on a Generative Adversarial Network

Unexpected anomalies pose significant risks to the health and security of intelligent manufacturing systems. This paper proposes a generative adversarial network (GAN)-based anomaly detection framework specifically for monitoring robotic manipulator operation using a side-channel energy auditing mechanism. To tackle the limitation arising from the lack of labeled data, the GAN model is trained by a semi-supervised learning approach that identifies anomalies during online operations as outliers. The overfitting is purposely utilized during the model training to enlarge the difference between normal energy consumption patterns used for training and anomalous profiles in real-time testing. In addition, the GAN model is modified to use multiple discriminators to analyze the individual energy profile associated with each joint or motor. The anomaly is detected by evaluating the mean and standard deviation values of anomaly scores' distribution, and both values are continuously updated by Welford's algorithm in real time to take into account the effect of environmental variations during operations. The detection performance on our custom dataset demonstrates the feasibility of the proposed pipeline. Specifically, for physical attacks, the framework can achieve an accuracy of approximately 0.93 for instant-wise detection and 0.84 for event-wise detection.

Transformer-based autoencoder with ID constraint for unsupervised anomalous sound detection

Unsupervised anomalous sound detection (ASD) aims to detect unknown anomalous sounds of devices when only normal sound data is available. The autoencoder (AE) and self-supervised learning based methods are two mainstream methods. However, the AE-based methods could be limited as the feature learned from normal sounds can also fit with anomalous sounds, reducing the ability of the model in detecting anomalies from sound. The self-supervised methods are not always stable and perform differently, even for machines of the same type. In addition, the anomalous sound may be short-lived, making it even harder to distinguish from normal sound. This paper proposes an ID-constrained Transformer-based autoencoder (IDC-TransAE) architecture with weighted anomaly score computation for unsupervised ASD. Machine ID is employed to constrain the latent space of the Transformer-based autoencoder (TransAE) by introducing a simple ID classifier to learn the difference in the distribution for the same machine type and enhance the ability of the model in distinguishing anomalous sound. Moreover, weighted anomaly score computation is introduced to highlight the anomaly scores of anomalous events that only appear for a short time. Experiments performed on DCASE 2020 Challenge Task2 development dataset demonstrate the effectiveness and superiority of our proposed method.

Wayside Railway Switch and Crossing Monitoring Using Isolation Forest Anomaly Scores

Railway switch and crossing (S&C) systems have complicated moving structures compared with regular rail. They require multiple components that vary in complexity. The complexity of railway S&C, together with the fact that they are discontinuous points of the system, makes them vulnerable to defects such as squats. A squat on the switching rail could potentially cause rail breakage and lead to catastrophic results, such as derailment. In this study, a method based on anomaly scoring was investigated to estimate the status of an S&C system with respect to squat defects. The proposed method was tested in a real environment under controlled measurement sequences. The results show that the methods can differ between an S&C with squats and another one without them.

ASE: Anomaly scoring based ensemble learning for highly imbalanced datasets

Nowadays, many classification algorithms have been applied to various industries to help them work out their problems met in real-life scenarios. However, in many binary classification tasks, samples in the minority class only make up a small part of all instances, which leads to the datasets we get usually suffer from high imbalance ratio. Existing models sometimes treat minority classes as noise or ignore them as outliers encountering data skewing. In order to solve this problem, we propose a bagging ensemble learning framework ASE (Anomaly Scoring Based Ensemble Learning). This framework has a scoring system based on anomaly detection algorithms which can guide the resampling strategy by divided samples in the majority class into subspaces. Then specific number of instances will be under-sampled from each subspace to construct subsets by combining with the minority class. And we calculate the weights of base classifiers trained by the subsets according to the classification result of the anomaly detection model and the statistics of the subspaces. Experiments have been conducted which show that our ensemble learning model can dramatically improve the performance of base classifiers and is more efficient than other existing methods under a wide range of imbalance ratio, data scale and data dimension. ASE can be combined with various classifiers and every part of our framework has been proved to be reasonable and necessary.

A Feature Memory Rearrangement Network for Visual Inspection of Textured Surface Defects Toward Edge Intelligent Manufacturing

Recent advances in the industrial inspection of textured surfaces-in the form of visual inspection-have made such inspections possible for efficient, flexible manufacturing systems. We propose an unsupervised feature memory rearrangement network (FMR-Net) to accurately detect various textural defects simultaneously. Consistent with mainstream methods, we adopt the idea of background reconstruction; however, we innovatively utilize artificial synthetic defects to enable the model to recognize anomalies, while traditional wisdom relies only on defect-free samples. First, we employ an encoding module to obtain multiscale features of the textured surface. Subsequently, a contrastive-learning-based memory feature module (CMFM) is proposed to obtain discriminative representations and construct a normal feature memory bank in the latent space, which can be employed as a substitute for defects and fast anomaly scores at the patch level. Next, a novel global feature rearrangement module (GFRM) is proposed to further suppress the reconstruction of residual defects. Finally, a decoding module utilizes the restored features to reconstruct the normal texture background. In addition, to improve inspection performance, a two-phase training strategy is utilized for accurate defect restoration refinement, and we exploit a multimodal inspection method to achieve noise-robust defect localization. We verify our method through extensive experiments and test its practical deployment in collaborative edge--cloud intelligent manufacturing scenarios by means of a multilevel detection method, demonstrating that FMR-Net exhibits state-of-the-art inspection accuracy and shows great potential for use in edge-computing-enabled smart industries.

An efficient two-stage pipeline model with filtering algorithm for mislabeled malware detection

Most malware detectors rely on machine-learning approaches. However, new malware samples are growing very fast day by day, and their labeling is very expensive. Consequently, companies cannot label each sample submitted for analysis. Instead, they label certain samples and treat the remaining as benign. Unfortunately, malicious samples are mislabeled and become noisy inputs. Nevertheless, the accuracy of the detection model is diminished by the bias introduced by mislabeling. Even obfuscation and adversarial attacks make malware detection difficult sometimes. These challenges motivate us to propose a novel two-stage pipeline technique based on two distinct sample behavior specifiers. First, using dynamic analysis tools, system calls, and application programming interface (API) call traces of samples are collected. Dynamic analysis is selected since it resolves the obfuscation issue significantly compared to static analysis. Second, API and system call traces are translated into a bag of n-grams or a sequence of anomaly scores. Then, the struc2vec algorithm is used to select the best n-gram features. The sequence of anomaly scores is transformed into six meta-features. A two-stage pipeline is proposed to detect the mislabeled malware and protect the host from Black-Box attacks. Each pipeline stage has a filtering algorithm for identifying the real label of any mislabeled malware. Experimental results on the VirusTotal and Ember datasets indicate that our proposed model achieves a 0.79 Matthews Correlation Coefficient (MCC) and a 0.62 MCC for classifying over 18,000 mislabeled malware samples, respectively. Also, our proposed model performs better in comparison to the existing models.

Machine learning model for detecting hydrogen leakage from hydrogen pipeline using physical modeling

Hydrogen pipelines (HPL) are one of the hydrogen transportation systems for realizing a hydrogen society. Hydrogen leakage from HPL is a challenge because hydrogen has a wide flammable range and low minimum ignition energy. Thus, hydrogen leakage from the HPL must be rapidly detected, and appropriate actions should be taken. Leakage detection is important for safe operation of HPL. The basic leakage detection method for HPL involves monitoring the pressure and flow rate values of the sensors. However, in some cases, it is difficult to distinguish between non-leakage and leakage conditions using this method. In this study, we focus on a leakage detection method using machine learning (ML) based on the relationship between pressure and flow rate data. There are two challenges in applying the ML- based leak detection method to an HPL. First, there are insufficient operational data for ML during the process- design stage. Secondly, it is difficult to obtain the pressure and flow rate behaviors during hydrogen leakage because leakage does not occur frequently. Consequently, this study employed an unsupervised ML method based on data simulated using a physical model of the HPL. First, a physical model of the HPL (HPL model) was constructed, and an ML model for leak detection was constructed based on the data simulated by the HPL model. The leak detection capability of the ML model was verified by comparing the anomaly scores of the non-leakage and leakage conditions. From the results, the ML model can distinguish between non-leakage and leakage behaviors and identify leakage points under certain conditions. This method can contribute to the optimization of the sensors required for leak detection during the process design stage.

Failure Prediction of Hard Disk Drives in Redundant Arrays Using Disk-Level Performance Metrics

Although many hard drive failure prediction methods utilize Self-Monitoring Analysis and Reporting Technology (SMART) features, they are not collected in IT systems with demanding performance requirements to meet their specification. We present a novel data-driven method for the prediction utilizing disk-level performance metrics collected by Redundant Array of Independent Disk (RAID) controllers instead of SMART features. The proposed method computes relational anomaly scores leveraging logical relationships of Hard Disk Drives (HDDs) based on RAID configuration for better failure prediction. In addition, it further utilizes error codes from HDDs to filter out false positives. We evaluate the proposed method on a real-world dataset collected for this study from 881 disks used in disk arrays of RAID-6 and 1660 disks used in disk arrays of RAID-10 in a data center. The results show consistent performance improvement by the logical relationships and error-code-based filtering. In addition, seven out of nine failures are predicted one day before the failure at the latest. This result suggests that the proposed method provides plenty of time for HDD replacement before a failure occurs.

Machine Learning-Based Reliability Improvement of Ambient Mode Extraction for Smart Grid Utilizing Isolation Forest

Reliable and accurate extraction of ambient modes is an essential means of assessing the safety and stability of smart grids. While previous works have mainly concentrated on the introduction of novel identification tools. In this paper, the isolation forest ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">iForest</i> ) was presented to build an intelligent ambient mode extraction scheme with high reliability, which aims at detecting and eliminating abnormal ambient modes from extracted results. As the core technology of the proposed intelligent scheme, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">iForest</i> exploits the concept of isolation and intends to combine the features of the outliers, which are ‘few’ and ‘different’, with the technique of ensemble learning to isolate them precisely. And the abnormal modes can be eliminated according to their anomaly scores in the proposed intelligent ambient mode extraction scheme. The evaluation of the proposed scheme is carried out through the database constructed by an IEEE 16-generator system and a real power system, the results of which indicate that the proposed scheme is intelligent enough to deal with a large amount of data with high accuracy and improve the reliability of existing identification tools.

Realizing Emergent Topics in Twitter by Detection of Abnormality Links

In the present paper, we demonstrated the probability model which may capture the usual mentioning behavior of an end user consisting of both the number of mentions for every post and the frequency of users occurring in the mentions. Subsequently, to compute the anomaly of future user behavior this model is needed. While using the proposed probability model, we can quantitatively compute the originality or probable effect of a post resembled in the mentioned behavior of the end user. We aggregate all the anomaly scores obtained in this way above countless end users. The efficiency of the proposed method is demonstrated on four data sets we have obtained through Twitter. We demonstrated that the mention-anomaly-based method combined with the TFIDF method can detect the emergence of a new topic at least as quickly as text-anomaly-based counterparts.

Unsupervised anomaly appraisal of cleft faces using a StyleGAN2-based model adaptation technique.

A novel machine learning framework that is able to consistently detect, localize, and measure the severity of human congenital cleft lip anomalies is introduced. The ultimate goal is to fill an important clinical void: to provide an objective and clinically feasible method of gauging baseline facial deformity and the change obtained through reconstructive surgical intervention. The proposed method first employs the StyleGAN2 generative adversarial network with model adaptation to produce a normalized transformation of 125 faces, and then uses a pixel-wise subtraction approach to assess the difference between all baseline images and their normalized counterparts (a proxy for severity of deformity). The pipeline of the proposed framework consists of the following steps: image preprocessing, face normalization, color transformation, heat-map generation, morphological erosion, and abnormality scoring. Heatmaps that finely discern anatomic anomalies visually corroborate the generated scores. The proposed framework is validated through computer simulations as well as by comparison of machine-generated versus human ratings of facial images. The anomaly scores yielded by the proposed computer model correlate closely with human ratings, with a calculated Pearson's r score of 0.89. The proposed pixel-wise measurement technique is shown to more closely mirror human ratings of cleft faces than two other existing, state-of-the-art image quality metrics (Learned Perceptual Image Patch Similarity and Structural Similarity Index). The proposed model may represent a new standard for objective, automated, and real-time clinical measurement of faces affected by congenital cleft deformity.

Energy consumption auditing based on a generative adversarial network for anomaly detection of robotic manipulators

This paper presents a generative adversarial network (GAN)-based energy auditing and anomaly detection framework for the health and security monitoring of robotic manipulators. The system features two key components: a side-channel measurement of the energy consumption and a GAN model to learn the data patterns during normal operations only. Anomalies in energy consumption during online testing are then discovered as outliers to the normal data by scrutinizing the anomaly score produced by the discriminator of the GAN model. This semi-supervised network circumvents the need for vast data labeling and exploits the overfitting during the training stage, hence magnifying the difference between the normal patterns learned in training and the unobserved abnormal energy profiles in testing. Moreover, considering the physical aspects of the robotic manipulator, a modified GAN architecture containing multiple discriminators to account for distinct contributions from individual joints/motors. A dynamic thresholding approach that continuously updates the statistical characteristics of the anomaly score is also developed to detect outliers and mitigate the effect of environmental variations during system operation. The proposed framework is tested on the custom dataset for performance evaluation. The results demonstrate the feasibility of the proposed method, especially for detecting physical attacks, which achieves an accuracy of approximately 0.93 in instant-level detection and over 0.84 in event-level detection.

Searching Density-Increasing Path to Local Density Peaks for Unsupervised Anomaly Detection

Unsupervised anomaly detection (AD) is a challenging problem in the data mining community. Clustering-based AD methods aim to group normal data points into clusters and then regard a point belonging to none of the clusters as an anomaly. However, they may suffer from the problems of unknown cluster numbers and arbitrary cluster shapes. This paper presents a novel clustering-based AD method named Density-increasing Path (DIP) to tackle these challenges. DIP searches a path for each data point. The path starts at the data point itself, passes through several points with monotonically increasing densities, and ends at a density peak. Further, DIP defines the climbing difficulty of each path by combining the distance and density increment of each step along the path, which can be regarded as the anomaly score of the path starting point. DIP can adaptively decide the number of peaks to address the challenge of unknown cluster numbers. Since DIP requires the path to pass several points rather than directly reaching the peak, it handles arbitrary cluster shapes. We also propose the ensemble DIP to improve prediction accuracy. The experimental results on four synthetic datasets and eleven real-world benchmarks demonstrate that DIP outperforms existing methods.

Time-Frequency Domain Monitoring Method for the Fault of HTS HVDC Systems Based on AI Classifiers

Since high-temperature superconductor (HTS) based HVDC systems are affected by temperature and cooling system performance, noise, the monitoring system which can classify the temporal disturbance and anomalies must be developed to ensure stable system operation. To solve this, it needs anomaly detection that accounts for system sensitivity and is able to continuously manage the response to disturbances. In this paper, we propose an integrated solution for fault severity diagnosis and anomaly detection colorred with AI based reflectometry. Anomalies such as the quench phenomenon can be detected through the anomaly score based on the reconstruction error calculated through the proposed autoencoder(AE) method. After anomaly detection, a fault classification algorithm based on the convolutional neural network (CNN) using a 2D image of a converted reflected signal through the proposed image processing was conducted. Based on the signal acquired from the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$7 m$</tex-math></inline-formula> length 1st generation 22.9 kV/50 MVA HTS cable, PSCAD simulation was utilized to construct a long-distance line model and verified the performance of the proposed algorithm.

Meta-survey on outlier and anomaly detection

The impact of outliers and anomalies on model estimation and data processing is of paramount importance, as evidenced by the extensive body of research spanning various fields over several decades: thousands of research papers have been published on the subject. As a consequence, numerous reviews, surveys, and textbooks have sought to summarize the existing literature, encompassing a wide range of methods from both the statistical and data mining communities. While these endeavors to organize and summarize the research are invaluable, they face inherent challenges due to the pervasive nature of outliers and anomalies in all data-intensive applications, irrespective of the specific application field or scientific discipline. As a result, the resulting collection of papers remains voluminous and somewhat heterogeneous.To address the need for knowledge organization in this domain, this paper implements the first systematic meta-survey of general surveys and reviews on outlier and anomaly detection. Employing a classical systematic survey approach, the study collects nearly 500 papers using two specialized scientific search engines. From this comprehensive collection, a subset of 56 papers that claim to be general surveys on outlier detection is selected using a snowball search technique to enhance field coverage. A meticulous quality assessment phase further refines the selection to a subset of 25 high-quality general surveys.Using this curated collection, the paper investigates the evolution of the outlier detection field over a 20-year period, revealing emerging themes and methods. Furthermore, an analysis of the surveys sheds light on the survey writing practices adopted by scholars from different communities who have contributed to this field.Finally, the paper delves into several topics where consensus has emerged from the literature. These include taxonomies of outlier types, challenges posed by high-dimensional data, the importance of anomaly scores, the impact of learning conditions, difficulties in benchmarking, and the significance of neural networks. Non-consensual aspects are also discussed, particularly the distinction between local and global outliers and the challenges in organizing detection methods into meaningful taxonomies.

Ensemble anomaly score for video anomaly detection using denoise diffusion model and motion filters

Video anomaly detection is a crucial task that aims to differentiate between normal and abnormal events. The current mainstream approach involves constructing an anomaly score based on the reconstruction error from a prediction model trained on normal frame sequences. However, this approach is limited by its deterministic nature, which may cause the anomaly score to be sensitive to underlying noise in the video. To address this limitation, this paper proposes an ensemble anomaly score constructed using a series of stochastic reconstructions of the original prediction. Specifically, we introduce the denoise diffusion model as a perturbation-denoise tool. First, the original prediction undergoes a perturbation process through a diffusion process. Then, a denoise diffusion model trained on normal predictions is used to directly reconstruct a series of noise-free predictions from the perturbed versions with different noise levels. Finally, an ensemble of all the reconstruction errors is used to provide a more generic and regularized anomaly score. Furthermore, we introduce motion filters into the detection pipeline to improve the modeling accuracy of the image distribution. The proposed method is evaluated on public datasets, and experimental results demonstrate its effectiveness, particularly in detecting performance under out-of-distribution (OOD) conditions.

Fuzzy granular anomaly detection using Markov random walk

Fuzzy information granulation is an important mathematical model in the theory of granular computing that can effectively handle fuzzy or uncertain information. To address the deficiencies of existing anomaly detection techniques that are mainly suitable for certainty data, this study proposes an anomaly detection method based on fuzzy granules. This method uses the fuzzy information granulation model as a unified framework to gradually develop a fuzzy granular anomaly detection method for calculating the anomaly score of each sample. First, the fuzzy granular distance is defined by fusing single-attribute fuzzy granules to represent the dissimilarity between two samples. Second, a matrix is constructed using the fuzzy granular distance between the samples as the state transition matrix of the Markov random walk. Anomalies in the dataset are then detected using the stationary distribution generated by iterative calculations. Finally, the fuzzy granular anomaly score for each object is obtained by normalizing the stationary distribution. Experiments are conducted on public datasets to compare the proposed method with some state-of-the-art anomaly detection methods. The results indicate that the proposed method is effective. The code is publicly available online at https://github.com/BElloney/FGAS.

RoSAS: Deep semi-supervised anomaly detection with contamination-resilient continuous supervision

Semi-supervised anomaly detection methods leverage a few anomaly examples to yield drastically improved performance compared to unsupervised models. However, they still suffer from two limitations: 1) unlabeled anomalies (i.e., anomaly contamination) may mislead the learning process when all the unlabeled data are employed as inliers for model training; 2) only discrete supervision information (such as binary or ordinal data labels) is exploited, which leads to suboptimal learning of anomaly scores that essentially take on a continuous distribution. Therefore, this paper proposes a novel semi-supervised anomaly detection method, which devises contamination-resilient continuous supervisory signals. Specifically, we propose a mass interpolation method to diffuse the abnormality of labeled anomalies, thereby creating new data samples labeled with continuous abnormal degrees. Meanwhile, the contaminated area can be covered by new data samples generated via combinations of data with correct labels. A feature learning-based objective is added to serve as an optimization constraint to regularize the network and further enhance the robustness w.r.t. anomaly contamination. Extensive experiments on 11 real-world datasets show that our approach significantly outperforms state-of-the-art competitors by 20%–30% in AUC-PR and obtains more robust and superior performance in settings with different anomaly contamination levels and varying numbers of labeled anomalies.

Improving Visual Defect Detection and Localization in Industrial Thermal Images Using Autoencoders.

Reliable functionality in anomaly detection in thermal image datasets is crucial for defect detection of industrial products. Nevertheless, achieving reliable functionality is challenging, especially when datasets are image sequences captured during equipment runtime with a smooth transition from healthy to defective images. This causes contamination of healthy training data with defective samples. Anomaly detection methods based on autoencoders are susceptible to a slight violation of a clean training dataset and lead to challenging threshold determination for sample classification. This paper indicates that combining anomaly scores leads to better threshold determination that effectively separates healthy and defective data. Our research results show that our approach helps to overcome these challenges. The autoencoder models in our research are trained with healthy images optimizing two loss functions: mean squared error (MSE) and structural similarity index measure (SSIM). Anomaly score outputs are used for classification. Three anomaly scores are applied: MSE, SSIM, and kernel density estimation (KDE). The proposed method is trained and tested on the 32 × 32-sized thermal images, including one contaminated dataset. The model achieved the following average accuracies across the datasets: MSE, 95.33%; SSIM, 88.37%; and KDE, 92.81%. Using a combination of anomaly scores could assist in solving a low classification accuracy. The use of KDE improves performance when healthy training data are contaminated. The MSE+ and SSIM+ methods, as well as two parameters to control quantitative anomaly localization using SSIM, are introduced.

On confidence computation and calibration of deep support vector data description

Deep support vector data description (DeSVDD) is an emerging anomaly detection method based on the deep learning methodology. However, few studies take the confidence of DeSVDD predictions into account so that the present DeSVDD models cannot indicate the reliability degree of the anomaly detection results. In this paper, we enrich the theory of DeSVDD by building the model confidence definition and developing the corresponding calibration strategy. For one thing, by revisiting the methodology of DeSVDD-based anomaly detection, the confidence of detection results is presented to indicate if the prediction of DeSVDD is reliable. For another, we propose a modified power T-scaling strategy to smooth the anomaly scores of DeSVDD model and improve its calibration performance without changing the original detection results. Six open experiment datasets are used to perform the method testing and the experimental results confirm the effectiveness of our proposed calibration strategy.