Showing papers on "Outlier published in 2023"

Local Temporal Pattern and Data Augmentation for Spotting Micro-Expressions

Abstract: Micro-expressions (MEs) are very important nonverbal communication clues. However, due to their local and short nature, spotting them is challenging. In this article, we address this problem by using a dedicated local and temporal pattern (LTP) of facial movement. This pattern has a specific shape (an S-pattern) when MEs are displayed. Thus, by using a classic classification algorithm (SVM), MEs can be distinguished from other facial movements. We also propose a global final fusion analysis covering the whole face to improve the distinction between ME (local) and head (global) movements. However, the learning of S-patterns is limited by the small number of ME databases and the low volume of ME samples. Hammerstein models (HMs) are known to effectively approximate muscle movements. By approximating each S-pattern with an HM, we can both filter out outliers and generate new similar S-patterns. In this way, we augment the dataset for S-pattern training and improve the ability to differentiate MEs from other movements. The spotting results, performed in the CASMEI and CASMEII databases, show that our proposed LTP outperforms the most popular spotting method in terms of the F1-score. Adding a fusion process and data augmentation improves the spotting performance even further.

Sampled Memory-Event-Triggered Fuzzy Load Frequency Control for Wind Power Systems Subject to Outliers and Transmission Delays

Abstract: This study is devoted to event-triggered fuzzy load frequency control (LFC) for wind power systems (WPSs) with measurement outliers and transmission delays. Due to the integration of wind turbine (WT) with nonlinearity, the T-S fuzzy model of WPS is established for stability analysis and controller design. To mitigate the network burden, a new sampled memory-event-triggered mechanism (SMETM) related to historical system information is presented. It has the following two merits: 1) the utilization of continuous memory outputs over a given interval is useful to reduce the information loss in the period of samples and the redundant triggering events induced by disturbances and noises and 2) an extra upper constraint is added in the triggering condition to generate a new event only when the error signal belongs to a bounded range, thus, the false events caused by measurement outliers can be differentiated out and then be dropped. By representing the memory signal with transmission delay as a time-varying distributed delay term, a T-S fuzzy time-varying distributed delay system is built up to model the H∞ LFC WPS. With the help of the Lyapunov method and the integral inequality relying on distributed delay, some criteria are derived to solve the triggering matrix and fuzzy controllers. Finally, the merits of the proposed SMETM are tested by simulation results.

Tracking Control Under Round-Robin Scheduling: Handling Impulsive Transmission Outliers

Abstract: In this article, the tracking control problem is investigated for a type of linear networked systems subject to the round-Robin (RR) protocol scheduling and impulsive transmission outliers (ITOs). The communication between the controller and sensors is implemented through a shared network, on which the signal transmissions are scheduled by the RR protocol. The considered ITOs are modeled by a sequence of impulsive signals whose amplitudes (i.e., the norms of all impulsive signals) and interval lengths (i.e., the duration between all adjacent impulsive signals) are greater than two known thresholds, respectively. The occurrence moment for each ITO is first examined by using a certain outlier detection approach, and then a novel parameter-dependent tracking controller is proposed to protect the tracking performance from ITOs by removing the "harmful" signals (i.e., the transmitted signals contaminated by ITOs). Sufficient conditions are presented to ensure the exponentially ultimate boundedness of the resulted tracking error, and the controller gain matrices are subsequently designed by solving a constrained optimization problem. Finally, a simulation example is provided to demonstrate the effectiveness of our developed outlier-resistant tracking control scheme.

Towards an Interpretable Autoencoder: A Decision-Tree-Based Autoencoder and its Application in Anomaly Detection

Abstract: The importance of understanding and explaining the associated classification results in the utilization of artificial intelligence (AI) in many different practical applications (e.g., cyber security and forensics) has contributed to the trend of moving away from black-box / opaque AI towards explainable AI (XAI). In this article, we propose the first interpretable autoencoder based on decision trees, which is designed to handle categorical data without the need to transform the data representation. Furthermore, our proposed interpretable autoencoder provides a natural explanation for experts in the application area. The experimental findings show that our proposed interpretable autoencoder is among the top-ranked anomaly detection algorithms, along with one-class Support Vector Machine (SVM) and Gaussian Mixture. More specifically, our proposal is on average 2% below the best Area Under the Curve (AUC) result and 3% over the other Average Precision scores, in comparison to One-class SVM, Isolation Forest, Local Outlier Factor, Elliptic Envelope, Gaussian Mixture Model, and eForest.

A Review of Techniques for Surface Electromyography Signal Quality Analysis

Abstract: Electromyography (EMG) signals are instrumental in a variety of applications including prosthetic control, muscle health assessment, rehabilitation, and workplace monitoring. Signal contaminants including noise, interference, and artifacts can degrade the quality of the EMG signal, leading to misinterpretation; therefore it is important to ensure that collected EMG signals are of sufficient quality prior to further analysis. A literature search was conducted to identify current approaches for detecting, identifying, and quantifying contaminants within surface EMG signals. We identified two main strategies: 1) bottom-up approaches for identifying specific and well-characterized contaminants and 2) top-down approaches for detecting anomalous EMG signals or outlier channels in high-density EMG arrays. The best type(s) of approach are dependent on the circumstances of data collection including the environment, the susceptibility of the application to contaminants, and the resilience of the application to contaminants. Further research is needed for assessing EMG with multiple simultaneous contaminants, identifying ground-truths for clean EMG data, and developing user-friendly and autonomous methods for EMG signal quality analysis.

Parameter-Free Loss for Class-Imbalanced Deep Learning in Image Classification

Abstract: Current state-of-the-art class-imbalanced loss functions for deep models require exhaustive tuning on hyperparameters for high model performance, resulting in low training efficiency and impracticality for nonexpert users. To tackle this issue, a parameter-free loss (PF-loss) function is proposed, which works for both binary and multiclass-imbalanced deep learning for image classification tasks. PF-loss provides three advantages: 1) training time is significantly reduced due to NO tuning on hyperparameter(s); 2) it dynamically pays more attention on minority classes (rather than outliers compared to the existing loss functions) with NO hyperparameters in the loss function; and 3) higher accuracy can be achieved since it adapts to the changes of data distribution in each mini-batch instead of the fixed hyperparameters in the existing methods during training, especially when the data are highly skewed. Experimental results on some classical image datasets with different imbalance ratios (IR, up to 200) show that PF-loss reduces the training time down to 1/148 of that spent by compared state-of-the-art losses and simultaneously achieves comparable or even higher accuracy in terms of both G-mean and area under receiver operating characteristic (ROC) curve (AUC) metrics, especially when the data are highly skewed.

Bayesian Inference for State-Space Models With Student-<i>t</i> Mixture Distributions

Abstract: This article proposes a robust Bayesian inference approach for linear state-space models with nonstationary and heavy-tailed noise for robust state estimation. The predicted distribution is modeled as the hierarchical Student- $t$ distribution, while the likelihood function is modified to the Student- $t$ mixture distribution. By learning the corresponding parameters online, informative components of the Student- $t$ mixture distribution are adapted to approximate the statistics of potential uncertainties. Then, the obstacle caused by the coupling of the updated parameters is eliminated by the variational Bayesian (VB) technique and fixed-point iterations. Discussions are provided to show the reasons for the achieved advantages analytically. Using the Newtonian tracking example and a three degree-of-freedom (DOF) hover system, we show that the proposed inference approach exhibits better performance compared with the existing method in the presence of modeling uncertainties and measurement outliers.

Stacking Ensemble Learning-Based Convolutional Gated Recurrent Neural Network for Diabetes Miletus

Abstract: Diabetes mellitus is a metabolic disease in which blood glucose levels rise as a result of pancreatic insulin production failure. It causes hyperglycemia and chronic multiorgan dysfunction, including blindness, renal failure, and cardiovascular disease, if left untreated. One of the essential checks that are needed to be performed frequently in Type 1 Diabetes Mellitus is a blood test, this procedure involves extracting blood quite frequently, which leads to subject discomfort increasing the possibility of infection when the procedure is often recurring. Existing methods used for diabetes classification have less classification accuracy and suffer from vanishing gradient problems, to overcome these issues, we proposed stacking ensemble learning-based convolutional gated recurrent neural network (CGRNN) Metamodel algorithm. Our proposed method initially performs outlier detection to remove outlier data, using the Gaussian distribution method, and the Box-cox method is used to correctly order the dataset. After the outliers’ detection, the missing values are replaced by the data’s mean rather than their elimination. In the stacking ensemble base model, multiple machine learning algorithms like Naïve Bayes, Bagging with random forest, and Adaboost Decision tree have been employed. CGRNN Meta model uses two hidden layers Long-Short-Time Memory (LSTM) and Gated Recurrent Unit (GRU) to calculate the weight matrix for diabetes prediction. Finally, the calculated weight matrix is passed to the softmax function in the output layer to produce the diabetes prediction results. By using LSTM-based CG-RNN, the mean square error (MSE) value is 0.016 and the obtained accuracy is 91.33%.

On Using Classification Datasets to Evaluate Graph Outlier Detection: Peculiar Observations and New Insights

Abstract: It is common practice of the outlier mining community to repurpose classification datasets toward evaluating various detection models. To that end, often a binary classification dataset is used, where samples from one of the classes are designated as the “inlier” samples, and the other class is substantially down-sampled to create the (ground-truth) “outlier” samples. Graph-level outlier detection (GLOD) is rarely studied but has many potentially influential real-world applications. In this study, we identify an intriguing issue with repurposing graph classification datasets for GLOD. We find that ROC-AUC performance of the models changes significantly (“flips” from high to very low, even worse than random) depending on which class is down-sampled. Interestingly, ROC-AUCs on these two variants approximately sum to 1 and their performance gap is amplified with increasing propagations for a certain family of propagation-based outlier detection models. We carefully study the graph embedding space produced by propagation-based models and find two driving factors: (1) disparity between within-class densities, which is amplified by propagation, and (2) overlapping support (mixing of embeddings) across classes. We also study other graph embedding methods and downstream outlier detectors, and we find that the intriguing “performance flip” issue still widely exists but which version of the down-sample achieves higher performance may vary. Thoughtful analysis over comprehensive results further deepens our understanding of the established issue. With this study, we aim at drawing attention to this (to our knowledge) previously unnoticed issue for the rarely studied GLOD problem, and specifically to the following questions: (1) Given the performance flip issue we identified, where one version of the down-sample often yields worse-than-random performance, is it appropriate to evaluate GLOD by average performance across all down-sampled versions when repurposing graph classification datasets? (2) Considering consistently observed performance flip issue across different graph embedding methods we studied, is it possible to design better graph embedding methods to overcome the issue? We conclude the article with our insights to these questions.

Performance Analysis of Robust Cooperative Positioning Based on GPS/UWB Integration for Connected Autonomous Vehicles

Abstract: The accurate position is a key requirement for autonomous vehicles. Although Global Navigation Satellite Systems (GNSS) are widely used in many applications, their performance is often disturbed, particularly in urban areas. Therefore, many studies consider multi-sensor integration and cooperative positioning (CP) approaches to provide additional degrees of freedom to address the shortcomings of GNSS. However, few studies adopted real-world datasets and internode ranging outliers within CP is left untouched, leading to unexpected challenges in practical applications. To address this, we propose a Robust Cooperative Positioning (RCP) scheme that augments the GPS with the Ultra-Wideband (UWB) system. A field experiment is conducted to generate a real-world dataset to evaluate the RCP scheme. Moreover, the analysis of the collected dataset enables us to optimise a simple but effective Robust Kalman Filter (RKF) to mitigate the influence of outlier measurements and improve the robustness of the proposed solution. Finally, a simulated dataset is derived from the real-world data to comprehensively study the performance of the proposed RCP method in urban canyon scenarios. Our results demonstrate that the proposed solution can crucially improve positioning performance when the number of visible GPS satellite is limited and is robust against various adverse effects in such environments.

Density variation-based background filtering algorithm for low-channel roadside lidar data

Abstract: Light Detection and Range (LiDAR) sensor is considered will be widely deployed in the roadside infrastructure if massive production in the near future, as it can extract High-Resolution Micro-level Traffic Data (HRMTD) which is a cornerstone in Intelligent Transportation Systems (ITS) applications. In the field application, background filtering is the first and foremost step to accelerate HRMTD extraction efficiency and improve extraction precision. In this paper, we proposed a novel background filtering algorithm based on density variation for low-channel roadside LiDAR. First, we segmented the detected area into small cubes and analyzed the character of LiDAR points in the detected area by calculating the density variation of the point cloud in continuous time. Second, we constructed an index to distinguish the road user passing area and removed outliers through the DBSCAN algorithm. Third, we excluded the LiDAR points that were not in the passing area. In the experiments, object points obtained percentage, background points excluded percentage, and effective points percentage were used to evaluate the accuracy of background filtering methods. Compared to the state-of-the-art methods, our algorithm has higher filtering accuracy and can perform well in complex sites in real-time. Besides, the proposed algorithm has the best stability, reflecting that the accuracy of the proposed methods does not decrease significantly as distance increases.

Robust PPG-based Mental Workload Assessment System using Wearable Devices

Abstract: Heart rate variability (HRV) has been used in assessing mental workload(MW) level. Compared with ECG, photoplethysmogram (PPG) provides convenient in assessing MW with wearable devices, which is more suitable for daily usage. However, PPG collected by smartwatches are prone to suffer from artifacts. Those signal corruptions cause invalid Inter-beat Intervals (IBI), making it challenging to evaluate the HRV feature. Hence, the PPG-based MW assessment system is difficult to obtain a sustainable and reliable assessment of MW. In this paper, we propose a pre- and post- processing technique, called outlier removal and uncertainty estimation, respectively, to reduce the negative influences of invalid IBIs. The proposed method helps to acquire accurate HRV features and evaluate the reliability of incoming IBIs, rejecting possibly misclassified data. We verified our approach in two open datasets, which are CLAS and MAUS. Experiment results show proposed method achieved higher accuracy (66.7% v.s. 74.2%) and lower variance (11.3% v.s. 10.8%) among users, which has comparable performance to an ECG-based MW system.

Isolation Forest Based Submodule Open-Circuit Fault Localization Method for Modular Multilevel Converters

Abstract: Fault localization is one of the most important issues for modular multilevel converters (MMCs) consisting of numerous switches. This article proposes an isolation forest (IF) based submodule (SM) switch open-circuit fault localization method for MMCs. Based on the continuous sampling SM capacitor voltages, a number of isolation trees (ITs) are produced to construct the IFs for MMCs. Through the comparison of continuous IFs’ outputs, the faulty SM can be effectively localized. The proposed IF-based fault localization method only requires SM capacitor voltages in the MMC to construct concise low-data-volume tree models, and uses sparsity and difference properties of outlier data to localize fault, and accordingly it simplifies calculation complexity. In addition, it does not require the MMC's mathematical models and manual setting of empirical thresholds. Simulation and experiment are conducted, and the results confirm the effectiveness of proposed method.

Integrated Depths for Partially Observed Functional Data

Abstract: Abstract Partially observed functional data are frequently encountered in applications and are the object of an increasing interest by the literature. We here address the problem of measuring the centrality of a datum in a partially observed functional sample. We propose an integrated functional depth for partially observed functional data, dealing with the very challenging case where partial observability can occur systematically on any observation of the functional dataset. In particular, differently from many techniques for partially observed functional data, we do not request that some functional datum is fully observed, nor we require that a common domain exist, where all of the functional data are recorded. Because of this, our proposal can also be used in those frequent situations where reconstructions methods and other techniques for partially observed functional data are inapplicable. By means of simulation studies, we demonstrate the very good performances of the proposed depth on finite samples. Our proposal enables the use of benchmark methods based on depths, originally introduced for fully observed data, in the case of partially observed functional data. This includes the functional boxplot, the outliergram and the depth versus depth classifiers. We illustrate our proposal on two case studies, the first concerning a problem of outlier detection in German electricity supply functions, the second regarding a classification problem with data obtained from medical imaging. Supplementary materials for this article are available online.

An Efficient Heterogeneous Edge-Cloud Learning Framework for Spectrum Data Compression

Abstract: Spectrum data compression with a high-rate compression and accurate reconstruction is of crucial importance for reducing the ultra-large data transmission from the edge sensors to the cloud for establishing high-quality spectrum maps. However, the current methods ignore the imbalanced edge-cloud computation resources and cannot tackle the outlier signals, resulting in significant challenges for achieving effective compression. Therefore, we develop an efficient heterogeneous edge-cloud learning framework. In the framework, paralleled methods compress normal data and outlier data distinctively based on their different structure information. Meanwhile, those methods are asymmetric for achieving low-cost compression at the edge and accurate reconstruction on the cloud. Based on the framework, we propose an outlier-processable attention-based asymmetric compression algorithm. A novel attention-based asymmetric convolutional neural network performs the normal data compression while a non-linear outlier compression algorithm realizes the outlier data compression. Compared with the state-of-the-art schemes in real-world settings, our proposed framework’s convergence speed increases by 120% . Meanwhile, our framework’s reconstruction accuracy increases by 68.42% under the interfered environments while maintaining superior compression speed and comprehensive performance. We also confirm our framework’s generalization ability to transfer among different tasks by deploying it under various spectrum environments.

A comprehensive platform for analyzing longitudinal multi-omics data

Abstract: Longitudinal bulk and single-cell omics data is increasingly generated for biological and clinical research but is challenging to analyze due to its many intrinsic types of variations. We present PALMO ( https://github.com/aifimmunology/PALMO ), a platform that contains five analytical modules to examine longitudinal bulk and single-cell multi-omics data from multiple perspectives, including decomposition of sources of variations within the data, collection of stable or variable features across timepoints and participants, identification of up- or down-regulated markers across timepoints of individual participants, and investigation on samples of same participants for possible outlier events. We have tested PALMO performance on a complex longitudinal multi-omics dataset of five data modalities on the same samples and six external datasets of diverse background. Both PALMO and our longitudinal multi-omics dataset can be valuable resources to the scientific community.

Variational Bayesian Inference for Robust Identification of PWARX Systems With Time-Varying Time-Delays

Abstract: This article presents a robust variational Bayesian (VB) algorithm for identifying piecewise autoregressive exogenous (PWARX) systems with time-varying time-delays. To alleviate the adverse effects caused by outliers, the probability distribution of noise is taken to follow a $t$ -distribution. Meanwhile, a solution strategy for more accurately classifying undecidable data points is proposed, and the hyperplanes used to split data are determined by a support vector machine (SVM). In addition, maximum-likelihood estimation (MLE) is adopted to re-estimate the unknown parameters through the classification results. The time-delay is regarded as a hidden variable and identified through the VB algorithm. The effectiveness of the proposed algorithm is illustrated by two simulation examples.

Rank-based testing for semiparametric VAR models: A measure transportation approach

Abstract: We develop a class of tests for semiparametric vector autoregressive (VAR) models with unspecified innovation densities, based on the recent measure-transportation-based concepts of multivariate {\it center-outward ranks} and {\it signs}. We show that these concepts, combined with Le Cam's asymptotic theory of statistical experiments, yield novel testing procedures, which (a)~are valid under a broad class of innovation densities (possibly non-elliptical, skewed, and/or with infinite moments), (b)~are optimal (locally asymptotically maximin or most stringent) at selected ones, and (c) are robust against additive outliers. In order to do so, we establish a H\' ajek asymptotic representation result, of independent interest, for a general class of center-outward rank-based serial statistics. As an illustration, we consider the problems of testing the absence of serial correlation in multiple-output and possibly non-linear regression (an extension of the classical Durbin-Watson problem) and the sequential identification of the order $p$ of a vector autoregressive (VAR($p$)) model. A Monte Carlo comparative study of our tests and their routinely-applied Gaussian competitors demonstrates the benefits (in terms of size, power, and robustness) of our methodology; these benefits are particularly significant in the presence of asymmetric and leptokurtic innovation densities. A real data application concludes the paper.