This work considers a task scheduling problem with deadline constraints in human-cyber-physical systems. To find its energy-efficient schedules in a short time, an autoencoder-embedded iterated local search algorithm ...
详细信息
This work considers a task scheduling problem with deadline constraints in human-cyber-physical systems. To find its energy-efficient schedules in a short time, an autoencoder-embedded iterated local search algorithm is proposed to solve it. Iterated local search is selected as a main scheduler. In order to handle real-time requirements and high computational load involved in the problem solution, a Long Short-Term Memory-based autoencoder model (LSTM-AE) is constructed to capture the relevant and complementary features of the considered problem. The model is used to find low-order and high-fit sub-solutions via unsupervised end-to-end learning, and generates promising solutions in an informative low-dimensional solution space. To further reduce computational burden, a two-stage optimization framework is constructed, which includes an off-line training phase and an online optimization one. The former trains LSTM-AE by using expert knowledge and historical data. The latter designs optimal resource allocation strategies to build a high-quality initial solution. Then, LSTM-AE-assisted local search operators are proposed and used to reform the initial solution and generate better ones. Various numerical experiments are performed to compare the proposed method with several classic heuristics and some recently-developed methods. The results show its superiority over them. Note to Practitioners-In human-cyber-physical systems, a task scheduling problem is usually solved by using heuristics due to the limited computational resources. Nevertheless, fast dispatching rules tend to perform poorly. Meta-heuristics can find a relatively high-quality schedule but are time-consuming, especially for a population-based algorithm that requires to evaluate a fitness function for many candidate solutions at each iteration. To balance computational burden and solution quality, our idea is to combine machine-learning methods with meta-heuristics. Specially, we integrate a long short-term
In complex industrial production environments, the efficacy of fault diagnostic techniques has become increasingly important and can enhance the reliability and safety of systems. In recent years, the discriminant loc...
详细信息
In complex industrial production environments, the efficacy of fault diagnostic techniques has become increasingly important and can enhance the reliability and safety of systems. In recent years, the discriminant locality-preserving projection (DLPP) algorithm has shown significant effectiveness in extracting meaningful features from complex industrial data. However, DLPP involves only the operation of projecting high-dimensional data onto a lower dimensional space, which is a one-way mapping process and lacks the verification of whether the low-dimensional data projected can accurately and effectively represent the original data. This might result in the loss of vital information in the original data, consequently limiting the performance of DLPP. In this article, we introduce a novel DLPP approach denoted as autoencoder-based DLPP (DLPP-AE), which is predicated upon the autoencoder. DLPP-AE establishes a bidirectional mapping process: in the encoding stage, DLPP serves as a mapping mechanism that transforms the initial high-dimensional data into a low-dimensional embedded representation;whereas in the decoding stage, the low-dimensional embedded data generated in the encoding stage are remapped back to its original high-dimensional form. This effectively resolves the issue of DLPP's inability to perform reverse validation on low-dimensional embeddings. To evaluate the performance of the proposed method, we conducted a comprehensive case study using three laboratory datasets and one real industrial dataset. The experimental results confirm the superior fault diagnosis capability of the DLPP-AE method.
Emergence of viruses cause unprecedented challenges and thus leading to wide-ranging consequences today. The world has faced massive disruptions like COVID-19 and continues to suffer in terms of public health and worl...
详细信息
Emergence of viruses cause unprecedented challenges and thus leading to wide-ranging consequences today. The world has faced massive disruptions like COVID-19 and continues to suffer in terms of public health and world economy. Fighting with this emergence of viruses and its reemergence plays a critical role in the health care industry. Identification of novel virus-drug associations is a vital step in drug discovery. Prediction and prioritization of novel virus-drug associations through computational approaches is an alternative and best choice considering the cost and risk of biological experiments. This study proposes a method, KR-AEVDA that relies on k-nearest neighbor based reliable negative sample selection and autoencoder based feature extraction to explore promising virus-drug associations for further experimental validation. The method analyzes complex relationships among drugs and viruses by investigating similarity and association data between drugs and viruses. It generates feature vectors from the similarity data, and reliable negative samples are extracted through an effective distance-based algorithm from the unlabeled samples in the dataset. Then high level features are extracted via an autoencoder and is fed to an ensemble classifier for inferring novel associations. Experimental results on three different datasets showed that KR-AEVDA reliably attained better performance than other stateof-the-art methods. Molecular docking is carried out between the top-predicted drugs and the crystal structure of the SARS-CoV-2's main protease to further validate the predictions. Case studies for SARS-CoV-2 illustrate the effectiveness of KR-AEVDA in identifying potential virus-drug associations.
The autoencoder (AE) is popular in Outlier Detection (OD) now due to its strong modeling ability. However, AE-based OD methods face the unexpected reconstruction problem: outliers are reconstructed with low errors, im...
详细信息
The autoencoder (AE) is popular in Outlier Detection (OD) now due to its strong modeling ability. However, AE-based OD methods face the unexpected reconstruction problem: outliers are reconstructed with low errors, impeding their distinction from inliers. This stems from two aspects. First, AE may overconfidently produce good reconstructions in regions where outliers or potential outliers exist while using the mean squared error. To address this, the aleatoric uncertainty was introduced to construct the Probabilistic autoencoder (PAE), and the Weighted Negative Log-Likelihood (WNLL) was proposed to enlarge the score disparity between inliers and outliers. Second, AE focuses on global modeling yet lacks the perception of local information. Therefore, the Mean-Shift Scoring (MSS) method was proposed to utilize the local relationship of data to reduce the false inliers caused by AE. Moreover, experiments on 32 real-world OD datasets proved the effectiveness of the proposed methods. The combination of WNLL and MSS achieved 45% relative performance improvement compared to the best baseline. In addition, MSS improved the detection performance of multiple AE-based outlier detectors by an average of 20%. The proposed methods have the potential to advance AE's development in OD.
Pulse pile-up is a problem in nuclear spectroscopy and nuclear reaction studies that occurs when two pulses overlap and distort each other,degrading the quality of energy and timing *** methods have been used for pile...
详细信息
Pulse pile-up is a problem in nuclear spectroscopy and nuclear reaction studies that occurs when two pulses overlap and distort each other,degrading the quality of energy and timing *** methods have been used for pile-up rejection,both digital and analogue,but some pile-up events may contain pulses of interest and need to be *** paper proposes a new method for reconstructing pile-up events acquired with a neutron detector array(NEDA)using an one-dimensional convolutional autoencoder(1D-CAE).The datasets for training and testing the 1D-CAE are created from data acquired from the *** new pile-up signal reconstruction method is evaluated from the point of view of how similar the reconstructed signals are to the original ***,it is analysed considering the result of the neutron-gamma discrimination based on charge comparison,comparing the result obtained from original and reconstructed signals.
This paper proposes a novel data-driven aeroelastic modeling method based on the autoencoder (AE) and the nonlinear state-space identification. This method allows high-dimensional flow field nonlinear features to be c...
详细信息
This paper proposes a novel data-driven aeroelastic modeling method based on the autoencoder (AE) and the nonlinear state-space identification. This method allows high-dimensional flow field nonlinear features to be characterized with high accuracy by lower-dimensional latent vectors, which facilitates the identification process and the generation of concise temporal models. The data-driven modeling initially employs a convolutional neural network autoencoder (CNN-AE) to reduce the dimensionality of pressure snapshots of unsteady flow fields obtained through high-fidelity numerical simulations. This process contributes to mapping the high-dimensional flow field to a low-dimensional latent space. Secondly, a temporal dynamics model of latent vectors is constructed using the state-space nonlinear identification. Subsequently, the temporal dynamics model is integrated with the decoder part of AE to reconstruct the temporal evolution of the unsteady transonic flow fields and aerodynamic forces. Finally, a nonlinear aeroelastic analysis is carried out by coupling the data-driven aerodynamic model and the structural model. In this paper, the effectiveness of this method in constructing a transonic data-driven model (DDM) is validated by the aerodynamic-structural coupling numerical example of the NACA0012 airfoil in a transonic regime. The results show that, in comparison to the linear flow modal decomposition method, the CNN-AE with nonlinear activation can utilize lower-dimensional latent vectors to represent the spatial structure of the flow field. Moreover, the resulting data-driven-based surrogate model is efficient and accurate in predicting the transonic flutter and limit-cycle oscillations (LCOs), proving a powerful tool for nonlinear aeroelastic analysis.
In this manuscript, we propose an innovative early warning Machine Learning-based model to identify potential threats to financial sustainability for non-financial companies. Unlike most state-of-the-art tools, whose ...
详细信息
In this manuscript, we propose an innovative early warning Machine Learning-based model to identify potential threats to financial sustainability for non-financial companies. Unlike most state-of-the-art tools, whose outcomes are often difficult to understand even for experts, our model provides an easily interpretable visualization of balance sheets, projecting each company in a bi-dimensional space according to an autoencoder-based dimensionality reduction matched with a Nearest-Neighbor-based default density estimation. In the resulting space, the distress zones, where the default intensity is high, appear as homogeneous clusters directly identified. Our empirical experiments provide evidence of the interpretability, forecasting ability, and robustness of the bi-dimensional space.
Dictionary representations and deep learning autoencoder (AE) models have proven effective in hyperspectral anomaly detection. Dictionary representations offer self-explanation but struggle with complex scenarios. Con...
详细信息
Dictionary representations and deep learning autoencoder (AE) models have proven effective in hyperspectral anomaly detection. Dictionary representations offer self-explanation but struggle with complex scenarios. Conversely, autoencoders can capture details in complex scenes but lack self-explanation. Complex scenarios often involve extensive spatial information, making its utilization crucial in hyperspectral anomaly detection. To effectively combine the advantages of both methods and address the insufficient use of spatial information, we propose an attention constrained low-rank and sparse autoencoder for hyperspectral anomaly detection. This model includes two encoders: an attention constrained low-rank autoencoder (AClrAE) trained with a background dictionary and incorporating a Global Self-Attention Module (GAM) to focus on global spatial information, resulting in improved background reconstruction;and an attention constrained sparse autoencoder (ACsAE) trained with an anomaly dictionary and incorporating a Local Self-Attention Module (LAM) to focus on local spatial information, resulting in enhanced anomaly reconstruction. Finally, to merge the detection results from both encoders, a nonlinear fusion scheme is employed. Experiments on multiple real and synthetic datasets demonstrate the effectiveness and feasibility of the proposed method.
Although existing graph self-supervised learning approaches have paid attention to the directed nature of networks, they have often overlooked the ubiquitous scale-free attributes. This oversight has resulted in a the...
详细信息
Although existing graph self-supervised learning approaches have paid attention to the directed nature of networks, they have often overlooked the ubiquitous scale-free attributes. This oversight has resulted in a theoretical gap in understanding graph self-supervised learning from the perspective of network structure. In this paper, we study the degree distribution characteristics of source and target nodes in directed scale-free networks, encompassing node and edge dimensions. Our theoretical analysis reveals the relationship between the average degree of nodes and their average in-degree and out-degree, which is instrumental in discerning positive and negative edges, as well as edge directionality. Here positive edges are the ones that exists in the original graph, and negative edges are the ones that not exists in the original graph. Furthermore, we uncover negative edges connecting to central nodes and positive edges to peripheral nodes to be less predictable. Based on these crucial theoretical insights, we propose DMGAE (Directed Masked Graph autoencoder), a novel representation learning method for directed scale-free networks that offers interpretability. The DMGAE method employs a weighted graph based on edges to replace the original graph structure. It integrates a masking approach based on the weight of the edges. Additionally, it incorporates an adaptive negative sampling method, edge decoder and a degree decoder based on the difference between the in- degree and out-degree of the node. This enhances the model's capability to learn edges and discern their directions. Empirical studies on extensive real-world network data show that, compared to the state-of-the-art methods, DMGAE not only generally has superior learning performance on directed networks, but also performs exceedingly well on undirected networks.
Intrusion detection (ID) gives security in network traffic or system activities monitors to detect suspicious activities, behavior, potential attacks, or unauthorized access. IDs are crucial in cybersecurity, as organ...
详细信息
Intrusion detection (ID) gives security in network traffic or system activities monitors to detect suspicious activities, behavior, potential attacks, or unauthorized access. IDs are crucial in cybersecurity, as organizations identify and respond to threats before they cause harm. The anomaly-based detection method is a popular and challenging research area in identifying new threats. So, this study focuses on developing an efficient network anomaly-based detection approach. It can establish a baseline for normal behavior and flag deviations from this baseline as potential threats. So, it can detect new or unknown attacks that deviate from standard traffic patterns. The study's main objective is to reduce the false positive rate and improve the class imbalance issues in the data. So, an optimized Deep Learning (DL) model is developed to detect new threats and reduce the false positive rate in the present ID systems. The DL model combines the Long Short-Term Memory (LSTM) with the autoencoder model, where the auto-encoder learns normal patterns, while LSTM handles sequential dependencies in the data. Moreover, the LSTM model performance is optimized using Particle Swarm Optimization (PSO). The performance of the DL model is evaluated with existing IDS methods and the results shows that the proposed model achieves maximum detection accuracy rate of 0.9989.
暂无评论