Input Data Samples Research Articles

ERA5 and ERA-Interim Data Processing for the GlobWat Global Hydrological Model

The reproducibility of computational hydrology is gaining attention among hydrologists. Reproducibility requires open and reusable code and data, allowing users to verify results and process new datasets. The creation of input files for global hydrological models (GHMs) requires complex high-resolution gridded dataset processing, limiting the model’s reproducibility to groups with advanced programming skills. GlobWat is one of these GHMs, which was developed by the Food and Agriculture Organization (FAO) to assess irrigation water use. Although the GlobWat code and sample input data are available, the methods for pre-processing model inputs are not available. Here, we present a set of open-source Python and YAML scripts within the Earth System Model Evaluation Tool (ESMValTool) that provide a formalized technique for developing and processing GlobWat model weather inputs. We demonstrate the use of these scripts with the ERA5 and ERA-Interim datasets from the European Centre for Medium-Range Weather Forecasts (ECMWF). To demonstrate the advantage of using these scripts, we ran the GlobWat model for 30 years for the entire world. The focus of the evaluation was on the Urmia Lake Basin in Iran. The validation of the model against the observed discharge in this basin showed that the combination of ERA5 and the De Bruin reference evaporation method yields the best GlobWat performance. Moreover, the scripts allowed us to examine the causes behind the differences in model outcomes.

Machine Learning with <i>Kay</i>

Abstract. Computational power is very important when training Deep Learning (DL) models with large amounts of data (Wooldridge, 2021). Hence, High-Performance Computing (HPC) can be leveraged to reduce computational cost, and the Irish Centre for High-End Computing (ICHEC) provides significant infrastructure and services for research and development to both academia and industry. A portion of ICHEC's HPC system has been allocated for institutional access, and this paper presents a case study of how to use Kay (Ireland's national supercomputer) in the remote sensing domain. Specifically, this study uses clusters of Kay Graphics Processing Units (GPUs) for training DL models to extract buildings from satellite imagery using a large number of input data samples.

Prediction of Building Energy Consumption Based on BP Neural Network

In order to solve the energy consumption hypothesis of large buildings, the energy consumption hypothesis based on the BP neural network is proposed. First, to study the system of statistical index of building energy consumption and the system of statistical reporting of energy consumption of civil construction. In addition, to establish reliable consumer authority control to ensure the security and management of the database. Second, based on an analysis of the mechanism by which the BP neural network operates, this article optimizes it and describes the structure of the neural network, which includes the number of network layers, the number of neurons in each layer, and the number of latent neuron layers. hidden neuron layers and hidden neurons. The maximum value method is used to normalize the input sample data; finally, the learning and training process of neural network is determined. Based on BP neural network theory, the energy consumption statistics platform and prediction system are established by using Delphi 6.0. These include functional modules such as basic building information management, building energy consumption information management, building energy consumption summary, energy presampling information management, and building energy consumption forecast; the collection of building energy consumption data is mainly completed by intelligent energy consumption monitoring sensor network system. Finally, the city’s building energy consumption information system conducts construction energy audits and analyzes the potential for energy savings. The results show that the hypothesis model determined by the BP neural network algorithm has an average error of 10.6% in predicting the construction energy consumption data, which is better than Matlab’s predicted result and the mean error is 12.6%. From this, it can be seen that the BP neural network algorithm can provide better predictions of building energy consumption.

A Cloud Computing-Based Intelligent Forecasting Method for Cross-Border E-Commerce Logistics Costs

Aiming at the problems of poor forecasting effect and low accuracy and low efficiency in current cross-border e-commerce logistics cost prediction methods, a cloud computing-based intelligent method for cross-border e-commerce logistics cost prediction is proposed. Analyze cloud computing concepts, characteristics, and service models, study cloud computing-related technologies, and train BP neural network algorithms based on BP neural network principles. The BP neural network structure is obtained by determining the number of neurons in the input layer, the number of neurons in the hidden layer, the number of neurons in the output layer, and the activation function of the neural network. Normalize the input data samples of the input layer, and select the initial weight, threshold, and learning rate parameters of the BP neural network to determine the momentum coefficient. This paper uses neural network model combined with Spark cloud computing platform to realize the intelligent prediction of cross-border e-commerce logistics cost. This method has good predictive ability. After a large amount of data input and output relationship training, it has obtained the most suitable model for prediction. The experimental results show that the cross-border e-commerce logistics cost prediction effect of the proposed method is good, and it can effectively improve the accuracy and efficiency of cross-border e-commerce logistics cost prediction.

Experimental comparison of the effectiveness of programs for minimizing systems of Boolean functions in the class of disjunctive normal forms

Objectives. Methods, algorithms and programs for solving problems of minimizing the DNF representations of Boolean functions are widely used in the design of digital systems to reduce the complexity (crystal area) of functional combinational blocks of digital systems placed into digital VLSI.The objective of the work is experimental comparison of domestic programs for minimizing Boolean functions in the DNF class included in the FLC-2 with two well-known foreign freely distributed programs for minimizing DNF known as Espresso IIC and ABC.Methods. Four sets sample of input data were used to compare the programs – there are widely known examples on which the effectiveness of the Espresso IIC program was tested and two sets of industrial examples from the practice of designing the logic circuits. Algorithms and programs for parallelization of calculations when separate functions of minimizing have been developed. Software tools for the application of joint minimization programs with separate minimization of functions are proposed.Results. The areas of preferred use and the execution time of programs for the source systems of functions (for minimization) characterized by large parameter values of dozens of arguments and functions, tens of thousands of elementary conjunctions are revealed. The efficiency of application of minimization programs for various forms of input data assignment is investigated – DNF, orthogonalized DNF, BDD (Binary Decision Diagrams) representations for systems of functions, truth tables and perfect DNF systems.Conclusion. The experimental results show the effectiveness of parallel programs – reducing the calculation time and increasing the dimensions of solved problems of separate minimization of Boolean function systems.

Fusion‐Learning‐Based Optimization: A Modified Metaheuristic Method for Lightweight High‐Performance Concrete Design

In order to build high‐quality concrete, it is imperative to know the raw materials in advance. It is possible to accurately predict the quality of concrete and the amount of raw materials used using machine learning‐enhanced methods. An automated process based on machine learning strategies is proposed in this paper for predicting the compressive strength of concrete. Fusion‐learning‐based optimization is used in the proposed approach to generate a strong learner by pooling support vector regression models. The SVR technique proposes an optimization method for finding the kernel radial basis function (RBF) parameters based on improving the innovative gunner algorithm (AIG). As a result of AIG's diverse solutions, local optima are effectively avoided. Therefore, the novelty of our research is that, in solving the uncertainty of predicted outputs based on integrated models, we use fusion‐learning‐based optimization to improve regression discrimination. We also collected a standard dataset to analyze the proposed algorithm, and subsequently, the dataset was designed from concrete laboratory tests on 244 samples, seven features, and three outputs. Different regression intensities are determined by correlation analysis of responses. Regression fusion is sufficiently accurate to estimate the number of desired outcomes examined based on the appropriate input data sample. The best quality concrete can be achieved with an error rate of less than 5%.

Early Diagnosis of Alzheimer’s Disease Based on Convolutional Neural Networks

Alzheimer’s disease (AD) is a neurodegenerative disorder, causing the most common dementia in the elderly peoples. The AD patients are rapidly increasing in each year and AD is sixth leading cause of death in USA. Magnetic resonance imaging (MRI) is the leading modality used for the diagnosis of AD. Deep learning based approaches have produced impressive results in this domain. The early diagnosis of AD depends on the efficient use of classification approach. To address this issue, this study proposes a system using two convolutional neural networks (CNN) based approaches for an early diagnosis of AD automatically. In the proposed system, we use segmented MRI scans. Input data samples of three classes include 110 normal control (NC), 110 mild cognitive impairment (MCI) and 105 AD subjects are used in this paper. The data is acquired from the ADNI database and gray matter (GM) images are obtained after the segmentation of MRI subjects which are used for the classification in the proposed models. The proposed approaches segregate among NC, MCI, and AD. While testing both methods applied on the segmented data samples, the highest performance results of the classification in terms of accuracy on NC vs. AD are 95.33% and 89.87%, respectively. The proposed methods distinguish between NC vs. MCI and MCI vs. AD patients with a classification accuracy of 90.74% and 86.69%. The experimental outcomes prove that both CNN-based frameworks produced state-of-the-art accurate results for testing.

Adaptive Attribute and Structure Subspace Clustering Network.

Deep self-expressiveness-based subspace clustering methods have demonstrated effectiveness. However, existing works only consider the attribute information to conduct the self-expressiveness, limiting the clustering performance. In this paper, we propose a novel adaptive attribute and structure subspace clustering network (AASSC-Net) to simultaneously consider the attribute and structure information in an adaptive graph fusion manner. Specifically, we first exploit an auto-encoder to represent input data samples with latent features for the construction of an attribute matrix. We also construct a mixed signed and symmetric structure matrix to capture the local geometric structure underlying data samples. Then, we perform self-expressiveness on the constructed attribute and structure matrices to learn their affinity graphs separately. Finally, we design a novel attention-based fusion module to adaptively leverage these two affinity graphs to construct a more discriminative affinity graph. Extensive experimental results on commonly used benchmark datasets demonstrate that our AASSC-Net significantly outperforms state-of-the-art methods. In addition, we conduct comprehensive ablation studies to discuss the effectiveness of the designed modules. The code is publicly available at https://github.com/ZhihaoPENG-CityU/AASSC-Net.

Probabilistic assessment of fatigue crack propagation with accounting for effects of crack retardation due to overloads

The paper presents a probabilistic approach to fatigue strength assessment based on the Wheeler model that provides the opportunity to account for crack retardation due to overload in a probabilistic formulation. The probabilistic assessment is performed by Monte Carlo simulations based on random sampling. It consists of running deterministic assessments a sufficiently large number of times with different combinations of randomly sampled input data in order to build a statistical picture of the output quantities. A computer code in Matlab was developed to obtain the prior assessment of the probability of fatigue failure of a pipeline structural component. The Bayesian inference is used to update the prior assessment failure probability assessments as soon data from in-service technical inspections become available

An efficient adversarial example generation algorithm based on an accelerated gradient iterative fast gradient

Most existing deep neural networks are susceptible to the influence of adversarial examples, which may cause them to output incorrect prediction results. An adversarial example is the addition of small noise disturbances to the input data samples, which will deceive the classification. Generally, these modifications are very small noise disturbances, which are not easily noticed by the human eye. However, most existing adversarial attacks achieve only low success rates in typical black-box settings, where the attackers have no prior knowledge about the model structures and/or model parameters. To tackle this problem, we propose an iterative algorithm based on an acceleration gradient to enhance the adversary attack. Our method accumulates the gradient of the loss function in each iteration and uses the next gradient information to influence the future gradient. We also introduce the scaling invariance principle of a deep neural network to optimize the input images for black-box attacks. In addition, to handle the drawbacks of a traditional iterative fast gradient sign method, we further present two gradient optimization methods. Experimental results on the ImageNet dataset show that our attack methods can achieve good transferability. In addition, those models obtained by integrated adversarial training with strong defense capability are also quite vulnerable to our black-box attack.

CATEGORY BIG DATA MODEL OF THE SOCIO-ECONOMIC SYSTEM

The socioeconomic systems in which the tasks of countering the legalization (laundering) of proceeds from crime and the financing of terrorism are solved are considered. An intermediate conclusion is presented and made that it is necessary to synthesize a problem-oriented control subsystem for such systems and tasks. The management tasks that distinguish such a data model in modern conditions are highlighted. It is proposed to use a categorical-theoretic approach for synthesizing a data model, which is free from the mathematical paradoxes of mathematical sets, and is also not limited by the size of the data sample from very large samples, such as big data. The purpose of modeling is to synthesize a mathematical model of big data from a variety of data sources of socioeconomic systems, which provides the classification of objects in interrelation (interaction) with other objects to identify such objects and relationships (interactions) that relate to legalization (laundering) proceeds from crime and financing of terrorism. The input samples are mathematically defined. A mathematical model of big data is synthesized in terms of category theory. The module provides the representation of input data samples in the form (format), which makes it possible to classify objects and categories of socioeconomic systems.

Seabed classification and source localization from ship spectrograms using deep learning models

An 18-layer residual neural network (ResNet-18) and a convolutional neural network (CNN) are used for source localization and seabed classification using broadband spectrograms obtained from ship-radiated noise. The propagation of the broadband noise from those ships of opportunity (SOO) contains significant information about the waveguide and seafloor. In this work, a ResNet-18 and a CNN are trained using multi-task learning to estimate the closest point of approach, ship speed, and seabed type. Training data are simulated using a range-independent normal modes model and several sediment profiles obtained by inversions in different regions. The input data samples correspond to 20 min long broadband spectrograms in the 300–1500 Hz band. During the training stage, data augmentation techniques are also applied to evaluate their effects on the predictions. The metrics used to evaluate performance are accuracy for seabed classification and root mean square error for source localization. Results demonstrate the ability of deep learning models to estimate source localization and seabed classification along with the impact of data augmentation techniques using SOO spectrograms. [Work supported by ONR, Contract No. N00014-19-C-2001.]

ROBUSTNESS AND ADVERSARIAL ATTACKS ON GENERATIVE MODELS

Since generative models rely on providing input data samples, it is essential to have a robust generative model capable of standing against adversarial attacks that can tamper with the model's output. This paper employs empirical analysis to examine the weaknesses of critical generative models like GANs and VAEs and additionally discovers the defense schemes. In a controlled environment created by accurately modeled adversarial trial data sets and time-sensitive analyses, we test and compare various confirmed adversarial training methods and defenses, such as implicit generative modeling and probabilistic adversarial robustness. Our results emphasize the difficulty of gaining complete robustness and suggest a way to deal with such attacks while preserving the model's accuracy. The analysis also reveals gaps in existing techniques, opening up possibilities for future research to improve the protection of generative models. This work will be valuable for the machine learning community in the future, as it contributes to discussing adversarial robustness and offers insights for researchers and practitioners.

Distributed associative memory network with memory refreshing loss

Despite recent progress in memory augmented neural network (MANN) research, associative memory networks with a single external memory still show limited performance on complex relational reasoning tasks. Especially the content-based addressable memory networks often fail to encode input data into rich enough representation for relational reasoning and this limits the relation modeling performance of MANN for long temporal sequence data. To address these problems, here we introduce a novel Distributed Associative Memory architecture (DAM) with Memory Refreshing Loss (MRL) which enhances the relation reasoning performance of MANN. Inspired by how the human brain works, our framework encodes data with distributed representation across multiple memory blocks and repeatedly refreshes the contents for enhanced memorization similar to the rehearsal process of the brain. For this procedure, we replace a single external memory with a set of multiple smaller associative memory blocks and update these sub-memory blocks simultaneously and independently for the distributed representation of input data. Moreover, we propose MRL which assists a task’s target objective while learning relational information existing in data. MRL enables MANN to reinforce an association between input data and task objective by reproducing stochastically sampled input data from stored memory contents. With this procedure, MANN further enriches the stored representations with relational information. In experiments, we apply our approaches to Differential Neural Computer (DNC), which is one of the representative content-based addressing memory models and achieves the state-of-the-art performance on both memorization and relational reasoning tasks.

Quantum Principal Component Analysis Only Achieves an Exponential Speedup Because of Its State Preparation Assumptions.

A central roadblock to analyzing quantum algorithms on quantum states is the lack of a comparable input model for classical algorithms. Inspired by recent work of the author [E. Tang, STOC 2019.], we introduce such a model, where we assume we can efficiently perform ℓ^{2}-norm samples of input data, a natural analog to quantum algorithms that assume efficient state preparation of classical data. Though this model produces less practical algorithms than the (stronger) standard model of classical computation, it captures versions of many of the features and nuances of quantum linear algebra algorithms. With this model, we describe classical analogs to Lloyd, Mohseni, and Rebentrost's quantum algorithms for principal component analysis [S. Lloyd, M. Mohseni, and P. Rebentrost, Nat. Phys. 10, 631 (2014).NPAHAX1745-247310.1038/nphys3029] and nearest-centroid clustering [S. Lloyd, M. Mohseni, and P. Rebentrost, Quantum algorithms for supervised and unsupervised machine learning]. Since they are only polynomially slower, these algorithms suggest that the exponential speedups of their quantum counterparts are simply an artifact of state preparation assumptions.

MULTI LABEL CLASSIFICATION FOR AN IMAGE USING CONVOLUTIONAL NEURAL NETWORKS

The machine learning has many capabilities one of them is classification. Classification employed in many contexts like telling hotel reviews good / bad, or inferring the image consists of dog, cat etc. As the data increases there is a need to organize it, for that purpose classification can be helpful. Modern classification problems involve the prediction of multiple labels simultaneously associated with a single instance known as "Multi Label Classification". In multi-label classification, each of the input data samples belongs to one or more than one classes or labels. The traditional binary and multi-class classification problems are the subset of the multi-label classification problem. In this paper we are implementing the multi label classification using CNN framework with keras libraries. Classification can be applied to different domain such as text, audio etc. In this paper we are applying classification for an image dataset.

Improvements to the Sliding Discrete Fourier Transform Algorithm [Tips & Tricks

This article presents two networks that improve upon the behavior and performance of previously published sliding discrete Fourier transform (SDFT) algorithms. The proposed networks are structurally simple, computationally efficient, guaranteed stable networks used for real-time sliding spectrum analysis. The first real-time network computes one spectral output sample, equal to a single-bin output of an N-point DFT, for each input signal sample. The second real-time network is frequency flexible, in that its analysis frequency can be any scalar value in the range of zero to one-half the input data sample rate measured in cycles per second.

An explainable one-dimensional convolutional neural networks based fault diagnosis method for building heating, ventilation and air conditioning systems

Due to the frequently changed outdoor weather conditions and indoor requirements, heating, ventilation and air conditioning (HVAC) experiences faulty operations inevitably throughout its lifespan. Therefore, it is important to monitor and diagnose HVAC fault operations. Recently, deep learning methods have attracted more attentions for their guarantee of better diagnosis performance under various system configurations and operating conditions. However, these methods are black-box models which though highly accurate for fault diagnosis but are extremely hard to explain. To overcome the disadvantage of poor interpretability of deep learning black-box models, this study therefore proposes a novel explainable deep learning based fault diagnosis method that is suitable for HVACs. To maintain HVAC operational information and variable locations of all chiller input data samples, proposed method is established with three characteristics: 1) the pooling layer is excluded, 2) the size of convolution filter kernel is set as 1, and 3) use softsign as an activation function. Considering the resulting impacts of HVAC faults on system operating variables, a new Absolute Gradient-weighted Class Activation Mapping (Grad-Absolute-CAM) method is proposed to visualize the fault diagnosis criteria and make the model explainable by providing the fault-discriminative information. The proposed method is validated using fault experimental dataset of a typical building HVAC system (i.e., chiller) from the ASHRAE research project 1043 (RP-1043). The fault diagnosis accuracy is over 98.5% for seven chiller faults. Results indicates that it is capable of interpreting the model work mechanism by activation feature maps and explaining the fault diagnosis criteria by Grad-Absolute-CAM.

A novel extreme learning machine based kNN classification method for dealing with big data

kNN algorithm, as an effective data mining technique, is always attended for supervised classification. On the other hand, the previously proposed kNN finding methods cannot be considered as efficient methods for dealing with big data. As there is daily generated and expanded big datasets on different online and offline servers, the efficient methods for such data must be introduced to find kNN. Moreover, massive amounts of data contain more noise and imperfection data samples that significantly increase the need for a robust kNN finding method. In this paper, a new fast and robust kNN finding framework is introduced to deal with the big datasets. In this method, a group of most relevant data samples to an input data sample are detected and the original kNN method is applied on them for finding the final nearest neighbors. The main goal of this method is dealing with the big datasets in an accurate, fast, and robust manner. Here, the training data samples of each label are grouped into some partitions based on the output of some mini-classifiers (i.e. ELM classifier). In fact, the behavior of the mini-classifiers is the basis of partitioning the training data samples. These mini-classifiers are trained using non-overlapping subsets of the training set in the form of each mini-classifier a subset. Here, an index is calculated for each partition to make the corresponding partition finding faster using a tree structure in which each partition index is fallen into a leaf. Then, outputs of the mini-classifiers for an input test sample are used to find the corresponding group of most relevant data samples to the input data sample on the tree. Experimental results indicate that the proposed method has better performance in most cases and comparable performance on other cases of original and noisy big data problems.

Accurate liver disease prediction system using convolutional neural network

Objectives: To introduce the technique which can ensure the accurate and reliable prediction of liver disease by adapting the deep learning technique. Methods: In this work Modified Convolutional Neural Network based Liver Disease Prediction System (MCNN-LDPS) is introduced for the accurate liver disease prediction outcome. In the proposed research work, Dimensionality reduction is carried out using Modified Principal Component Analysis. Optimal feature selection is carried out using Score based Artificial Fish Swarm Algorithm (SAFSA). In SAFSA algorithm, information gain and entropy values are taken as input values which proved accurate outcome. This research method has been analysed over Indian Liver patient dataset. Findings: The analysis of the research work proves that the proposed method MCNN-LDPS obtains better outcome in terms of increased accuracy, precision. Here comparison analysis proved that MCNN-LDPS obtains 4.05% increased accuracy, 21.23% F-measure, 4.22% precision and 34.26% recall. This research method has been compared with the existing Multi layer Perceptron Neural Network (MLPNN) for the performance analysis. Novelty: The major limitation of CNN is its inability to encode Orientational and relative spatial relationships, view angle. CNN do not encode the position and orientation of data. Lack of ability to be spatially invariant to the input data sample. This is resolved in this research work by combining the genetic algorithm with the CNN method. Keywords: Liver Disease Prediction; Feature Selection; Information Gain; Entropy; Convolutional neural network; Dimensionality Reduction