Activation functions play an essential role in converting the output of the artificial neural network into nonlinear results,since without this nonlinearity,the results of the network will be less accurate.Nonlinearity...Activation functions play an essential role in converting the output of the artificial neural network into nonlinear results,since without this nonlinearity,the results of the network will be less accurate.Nonlinearity is the mission of all nonlinear functions,except for polynomials.The activation function must be dif-ferentiable for backpropagation learning.This study’s objective is to determine the best activation functions for the approximation of each fractal image.Different results have been attained using Matlab and Visual Basic programs,which indi-cate that the bounded function is more helpful than other functions.The non-lin-earity of the activation function is important when using neural networks for coding fractal images because the coefficients of the Iterated Function System are different according to the different types of fractals.The most commonly cho-sen activation function is the sigmoidal function,which produces a positive value.Other functions,such as tansh or arctan,whose values can be positive or negative depending on the network input,tend to train neural networks faster.The coding speed of the fractal image is different depending on the appropriate activation function chosen for each fractal shape.In this paper,we have provided the appro-priate activation functions for each type of system of iterated functions that help the network to identify the transactions of the system.展开更多
Accurate prediction of the rate of penetration(ROP)is significant for drilling optimization.While the intelligent ROP prediction model based on fully connected neural networks(FNN)outperforms traditional ROP equations...Accurate prediction of the rate of penetration(ROP)is significant for drilling optimization.While the intelligent ROP prediction model based on fully connected neural networks(FNN)outperforms traditional ROP equations and machine learning algorithms,its lack of interpretability undermines its credibility.This study proposes a novel interpretation and characterization method for the FNN ROP prediction model using the Rectified Linear Unit(ReLU)activation function.By leveraging the derivative of the ReLU function,the FNN function calculation process is transformed into vector operations.The FNN model is linearly characterized through further simplification,enabling its interpretation and analysis.The proposed method is applied in ROP prediction scenarios using drilling data from three vertical wells in the Tarim Oilfield.The results demonstrate that the FNN ROP prediction model with ReLU as the activation function performs exceptionally well.The relative activation frequency curve of hidden layer neurons aids in analyzing the overfitting of the FNN ROP model and determining drilling data similarity.In the well sections with similar drilling data,averaging the weight parameters enables linear characterization of the FNN ROP prediction model,leading to the establishment of a corresponding linear representation equation.Furthermore,the quantitative analysis of each feature's influence on ROP facilitates the proposal of drilling parameter optimization schemes for the current well section.The established linear characterization equation exhibits high precision,strong stability,and adaptability through the application and validation across multiple well sections.展开更多
This paper presents a novel framework aimed at quantifying uncertainties associated with the 3D reconstruction of smoke from2Dimages.This approach reconstructs color and density fields from 2D images using Neural Radi...This paper presents a novel framework aimed at quantifying uncertainties associated with the 3D reconstruction of smoke from2Dimages.This approach reconstructs color and density fields from 2D images using Neural Radiance Field(NeRF)and improves image quality using frequency regularization.The NeRF model is obtained via joint training ofmultiple artificial neural networks,whereby the expectation and standard deviation of density fields and RGB values can be evaluated for each pixel.In addition,customized physics-informed neural network(PINN)with residual blocks and two-layer activation functions are utilized to input the density fields of the NeRF into Navier-Stokes equations and convection-diffusion equations to reconstruct the velocity field.The velocity uncertainties are also evaluated through ensemble learning.The effectiveness of the proposed algorithm is demonstrated through numerical examples.The presentmethod is an important step towards downstream tasks such as reliability analysis and robust optimization in engineering design.展开更多
The development of defect prediction plays a significant role in improving software quality. Such predictions are used to identify defective modules before the testing and to minimize the time and cost. The software w...The development of defect prediction plays a significant role in improving software quality. Such predictions are used to identify defective modules before the testing and to minimize the time and cost. The software with defects negatively impacts operational costs and finally affects customer satisfaction. Numerous approaches exist to predict software defects. However, the timely and accurate software bugs are the major challenging issues. To improve the timely and accurate software defect prediction, a novel technique called Nonparametric Statistical feature scaled QuAdratic regressive convolution Deep nEural Network (SQADEN) is introduced. The proposed SQADEN technique mainly includes two major processes namely metric or feature selection and classification. First, the SQADEN uses the nonparametric statistical Torgerson–Gower scaling technique for identifying the relevant software metrics by measuring the similarity using the dice coefficient. The feature selection process is used to minimize the time complexity of software fault prediction. With the selected metrics, software fault perdition with the help of the Quadratic Censored regressive convolution deep neural network-based classification. The deep learning classifier analyzes the training and testing samples using the contingency correlation coefficient. The softstep activation function is used to provide the final fault prediction results. To minimize the error, the Nelder–Mead method is applied to solve non-linear least-squares problems. Finally, accurate classification results with a minimum error are obtained at the output layer. Experimental evaluation is carried out with different quantitative metrics such as accuracy, precision, recall, F-measure, and time complexity. The analyzed results demonstrate the superior performance of our proposed SQADEN technique with maximum accuracy, sensitivity and specificity by 3%, 3%, 2% and 3% and minimum time and space by 13% and 15% when compared with the two state-of-the-art methods.展开更多
Background:Excessive heat exposure can lead to hyperthermia in humans,which impairs physical performance and disrupts cognitive function.While heat is a known physiological stressor,it is unclear how severe heat stres...Background:Excessive heat exposure can lead to hyperthermia in humans,which impairs physical performance and disrupts cognitive function.While heat is a known physiological stressor,it is unclear how severe heat stress affects brain physiology and function.Methods:Eleven healthy participants were subjected to heat stress from prolonged exercise or warm water immersion until their rectal temperatures(T_(re))attained 39.5℃,inducing exertional or passive hyperthermia,respectively.In a separate trial,blended ice was ingested before and during exercise as a cooling strategy.Data were compared to a control condition with seated rest(normothermic).Brain temperature(T_(br)),cerebral perfusion,and task-based brain activity were assessed using magnetic resonance imaging techniques.Results:T_(br)in motor cortex was found to be tightly regulated at rest(37.3℃±0.4℃(mean±SD))despite fluctuations in T_(re).With the development of hyperthermia,T_(br)increases and dovetails with the rising T_(re).Bilateral motor cortical activity was suppressed during high-intensity plantarflexion tasks,implying a reduced central motor drive in hyperthermic participants(T_(re)=38.5℃±0.1℃).Global gray matter perfusion and regional perfusion in sensorimotor cortex were reduced with passive hyperthermia.Executive function was poorer under a passive hyperthermic state,and this could relate to compromised visual processing as indicated by the reduced activation of left lateral-occipital cortex.Conversely,ingestion of blended ice before and during exercise alleviated the rise in both T_(re)and T_(bc)and mitigated heat-related neural perturbations.Conclusion:Severe heat exposure elevates T_(br),disrupts motor cortical activity and executive function,and this can lead to impairment of physical and cognitive performance.展开更多
Self-normalizing neural networks(SNN)regulate the activation and gradient flows through activation functions with the self-normalization property.As SNNs do not rely on norms computed from minibatches,they are more fr...Self-normalizing neural networks(SNN)regulate the activation and gradient flows through activation functions with the self-normalization property.As SNNs do not rely on norms computed from minibatches,they are more friendly to data parallelism,kernel fusion,and emerging architectures such as ReRAM-based accelerators.However,existing SNNs have mainly demonstrated their effectiveness on toy datasets and fall short in accuracy when dealing with large-scale tasks like ImageNet.They lack the strong normalization,regularization,and expression power required for wider,deeper models and larger-scale tasks.To enhance the normalization strength,this paper introduces a comprehensive and practical definition of the self-normalization property in terms of the stability and attractiveness of the statistical fixed points.It is comprehensive as it jointly considers all the fixed points used by existing studies:the first and second moment of forward activation and the expected Frobenius norm of backward gradient.The practicality comes from the analytical equations provided by our paper to assess the stability and attractiveness of each fixed point,which are derived from theoretical analysis of the forward and backward signals.The proposed definition is applied to a meta activation function inspired by prior research,leading to a stronger self-normalizing activation function named‘‘bi-scaled exponential linear unit with backward standardized’’(bSELU-BSTD).We provide both theoretical and empirical evidence to show that it is superior to existing studies.To enhance the regularization and expression power,we further propose scaled-Mixup and channel-wise scale&shift.With these three techniques,our approach achieves 75.23%top-1 accuracy on the ImageNet with Conv MobileNet V1,surpassing the performance of existing self-normalizing activation functions.To the best of our knowledge,this is the first SNN that achieves comparable accuracy to batch normalization on ImageNet.展开更多
Recently we have studied the rare earth ion-selective electrodes with active materials of the func-tional polymers and found that the process chosen for the functional polymers had an effect on the propertiesof gadoli...Recently we have studied the rare earth ion-selective electrodes with active materials of the func-tional polymers and found that the process chosen for the functional polymers had an effect on the propertiesof gadolinium ion selective electrode besides the effects of their structures.1.Effect of preparation process of the grafted polymers on the properties ofgadolinium ion selective electrodesThe electrode membranes which consist of functional polymers as active materials were prepared by re-action of gadolinium chloride with the radiation grafted clmer of acrlic acid and polystyrene of which展开更多
This paper describes our implementation of several neural networks built on a field programmable gate array (FPGA) and used to recognize a handwritten digit dataset—the Modified National Institute of Standards and Te...This paper describes our implementation of several neural networks built on a field programmable gate array (FPGA) and used to recognize a handwritten digit dataset—the Modified National Institute of Standards and Technology (MNIST) database. We also propose a novel hardware-friendly activation function called the dynamic Rectifid Linear Unit (ReLU)—D-ReLU function that achieves higher performance than traditional activation functions at no cost to accuracy. We built a 2-layer online training multilayer perceptron (MLP) neural network on an FPGA with varying data width. Reducing the data width from 8 to 4 bits only reduces prediction accuracy by 11%, but the FPGA area decreases by 41%. Compared to networks that use the sigmoid functions, our proposed D-ReLU function uses 24% - 41% less area with no loss to prediction accuracy. Further reducing the data width of the 3-layer networks from 8 to 4 bits, the prediction accuracies only decrease by 3% - 5%, with area being reduced by 9% - 28%. Moreover, FPGA solutions have 29 times faster execution time, even despite running at a 60× lower clock rate. Thus, FPGA implementations of neural networks offer a high-performance, low power alternative to traditional software methods, and our novel D-ReLU activation function offers additional improvements to performance and power saving.展开更多
In this paper, the multistability issue is discussed for delayed complex-valued recurrent neural networks with discontinuous real-imaginary-type activation functions. Based on a fixed theorem and stability definition,...In this paper, the multistability issue is discussed for delayed complex-valued recurrent neural networks with discontinuous real-imaginary-type activation functions. Based on a fixed theorem and stability definition, sufficient criteria are established for the existence and stability of multiple equilibria of complex-valued recurrent neural networks. The number of stable equilibria is larger than that of real-valued recurrent neural networks, which can be used to achieve high-capacity associative memories. One numerical example is provided to show the effectiveness and superiority of the presented results.展开更多
In this paper,the functional polymeric active materials were prepared by the grafting copolymerization and their structure and properties were studied.The results show that the structure and properties of these ac- ti...In this paper,the functional polymeric active materials were prepared by the grafting copolymerization and their structure and properties were studied.The results show that the structure and properties of these ac- tive materials have the relative large effects on the properties of gadolinium ion selective electrodes.展开更多
Respect for human rights and protection of human rights are significant rules in the Constitution of the People’s Republic of China.In 2015,judicial administration departments at all levels legally exercised their du...Respect for human rights and protection of human rights are significant rules in the Constitution of the People’s Republic of China.In 2015,judicial administration departments at all levels legally exercised their duties,implemented the principles and rules in constitution,and kept strengthening propagation of human rights through creation of contents and methods which had acquired great effects.What they have done contributes significantly toward the development of human rights in China.展开更多
An impact point prediction(IPP) guidance based on supervised learning is proposed to address the problem of precise guidance for the ballistic missile in high maneuver penetration condition.An accurate ballistic traje...An impact point prediction(IPP) guidance based on supervised learning is proposed to address the problem of precise guidance for the ballistic missile in high maneuver penetration condition.An accurate ballistic trajectory model is applied to generate training samples,and ablation experiments are conducted to determine the mapping relationship between the flight state and the impact point.At the same time,the impact point coordinates are decoupled to improve the prediction accuracy,and the sigmoid activation function is improved to ameliorate the prediction efficiency.Therefore,an IPP neural network model,which solves the contradiction between the accuracy and the speed of the IPP,is established.In view of the performance deviation of the divert control system,the mapping relationship between the guidance parameters and the impact deviation is analysed based on the variational principle.In addition,a fast iterative model of guidance parameters is designed for reference to the Newton iteration method,which solves the nonlinear strong coupling problem of the guidance parameter solution.Monte Carlo simulation results show that the prediction accuracy of the impact point is high,with a 3 σ prediction error of 4.5 m,and the guidance method is robust,with a 3 σ error of 7.5 m.On the STM32F407 singlechip microcomputer,a single IPP takes about 2.374 ms,and a single guidance solution takes about9.936 ms,which has a good real-time performance and a certain engineering application value.展开更多
Recently,deep learning has achieved remarkable results in fields that require human cognitive ability,learning ability,and reasoning ability.Activation functions are very important because they provide the ability of ...Recently,deep learning has achieved remarkable results in fields that require human cognitive ability,learning ability,and reasoning ability.Activation functions are very important because they provide the ability of artificial neural networks to learn complex patterns through nonlinearity.Various activation functions are being studied to solve problems such as vanishing gradients and dying nodes that may occur in the deep learning process.However,it takes a lot of time and effort for researchers to use the existing activation function in their research.Therefore,in this paper,we propose a universal activation function(UA)so that researchers can easily create and apply various activation functions and improve the performance of neural networks.UA can generate new types of activation functions as well as functions like traditional activation functions by properly adjusting three hyperparameters.The famous Convolutional Neural Network(CNN)and benchmark datasetwere used to evaluate the experimental performance of the UA proposed in this study.We compared the performance of the artificial neural network to which the traditional activation function is applied and the artificial neural network to which theUA is applied.In addition,we evaluated the performance of the new activation function generated by adjusting the hyperparameters of theUA.The experimental performance evaluation results showed that the classification performance of CNNs improved by up to 5%through the UA,although most of them showed similar performance to the traditional activation function.展开更多
Machine Learning(ML)and Deep Learning(DL)technologies are revolutionizing the medical domain,especially with Electrocardiogram(ECG),by providing new tools and techniques for diagnosing,treating,and preventing diseases...Machine Learning(ML)and Deep Learning(DL)technologies are revolutionizing the medical domain,especially with Electrocardiogram(ECG),by providing new tools and techniques for diagnosing,treating,and preventing diseases.However,DL architectures are computationally more demanding.In recent years,researchers have focused on combining the computationally less intensive portion of the DL architectures with ML approaches,say for example,combining the convolutional layer blocks of Convolution Neural Networks(CNNs)into ML algorithms such as Extreme Gradient Boosting(XGBoost)and K-Nearest Neighbor(KNN)resulting in CNN-XGBoost and CNN-KNN,respectively.However,these approaches are homogenous in the sense that they use a fixed Activation Function(AFs)in the sequence of convolution and pooling layers,thereby limiting the ability to capture unique features.Since various AFs are readily available and each could capture unique features,we propose a Convolutionbased Heterogeneous Activation Facility(CHAF)which uses multiple AFs in the convolution layer blocks,one for each block,with a motivation of extracting features in a better manner to improve the accuracy.The proposed CHAF approach is validated on PTB and shown to outperform the homogeneous approaches such as CNN-KNN and CNN-XGBoost.For PTB dataset,proposed CHAF-KNN has an accuracy of 99.55%and an F1 score of 99.68%in just 0.008 s,outperforming the state-of-the-art CNN-XGBoost which has an accuracy of 99.38%and an F1 score of 99.32%in 1.23 s.To validate the generality of the proposed CHAF,experiments were repeated on MIT-BIH dataset,and the proposed CHAF-KNN is shown to outperform CNN-KNN and CNN-XGBoost.展开更多
Renewable energy production plays a major role in satisfying electricity demand.Wind power conversion is one of the most popular renewable energy sources compared to other sources.Wind energy conversion has two major ...Renewable energy production plays a major role in satisfying electricity demand.Wind power conversion is one of the most popular renewable energy sources compared to other sources.Wind energy conversion has two major types of generators such as the Permanent Magnet Synchronous Generator(PMSG)and the Doubly Fed Induction Generator(DFIG).The maximum power tracking algo-rithm is a crucial controller,a wind energy conversion system for generating maximum power in different wind speed conditions.In this article,the DFIG wind energy conversion system was developed in Matrix Laboratory(MATLAB)and designed a machine learning(ML)algorithm for the rotor and grid side converter.The ML algorithm has been developed and trained in a MATLAB environment.There are two types of learning algorithms such as supervised and unsupervised learning.In this research supervised learning is used to power the neural networks and analysis is made for various hidden layers and activation functions.Simulation results are assessed to demonstrate the efficiency of the proposed system.展开更多
Multi-port converters are considered as exceeding earlier period decade owing to function in a combination of different energy sources in a single processing unit.Renewable energy sources are playing a significant rol...Multi-port converters are considered as exceeding earlier period decade owing to function in a combination of different energy sources in a single processing unit.Renewable energy sources are playing a significant role in the modern energy system with rapid development.In renewable sources like fuel combustion and solar energy,the generated voltages change due to their environmental changes.To develop energy resources,electric power generation involved huge awareness.The power and output voltages are plays important role in our work but it not considered in the existing system.For considering the power and voltage,Gaussian PI Controller-Maxpooling Deep Convolutional Neural Network Classifier(GPIC-MDCNNC)Model is introduced for the grid-connected renewable energy system.The input information is collected from two input sources.After that,input layer transfer information to hidden layer 1 fuzzy PI is employed for controlling voltage in GPIC-MDCNNC Model.Hidden layer 1 is transferred to hidden layer 2.Gaussian activation is employed for determining the output voltage with help of the controller.At last,the output layer offers the last value in GPIC-MDCNNC Model.The designed method was confirmed using one and multiple sources by stable and unpredictable input voltages.GPIC-MDCNNC Model increases the performance of grid-connected renewable energy systems by enhanced voltage value compared with state-of-the-art works.The control technique using GPIC-MDCNNC Model increases the dynamics of hybrid energy systems connected to the grid.展开更多
Ginkgo biloba resources in China are enormous. With the demand of the market, the preparation and application of flavonoids have become a current research hotspot. The main active substances in G. biloba leaves, flavo...Ginkgo biloba resources in China are enormous. With the demand of the market, the preparation and application of flavonoids have become a current research hotspot. The main active substances in G. biloba leaves, flavonoids, have various functional activities and are widely used in fields such as food, medicine, cosmetics, feed, etc. In this paper, the introduction, functional activity, extraction methods, and application research of flavonoids from G. biloba leaves are reviewed, and the development prospects of flavonoids from G. biloba leaves are expected.展开更多
In this study, the author will investigate and utilize advanced machine learning models related to two different methodologies to determine the best and most effective way to predict individuals with heart failure and...In this study, the author will investigate and utilize advanced machine learning models related to two different methodologies to determine the best and most effective way to predict individuals with heart failure and cardiovascular diseases. The first methodology involves a list of classification machine learning algorithms, and the second methodology involves the use of a deep learning algorithm known as MLP or Multilayer Perceptrons. Globally, hospitals are dealing with cases related to cardiovascular diseases and heart failure as they are major causes of death, not only for overweight individuals but also for those who do not adopt a healthy diet and lifestyle. Often, heart failures and cardiovascular diseases can be caused by many factors, including cardiomyopathy, high blood pressure, coronary heart disease, and heart inflammation [1]. Other factors, such as irregular shocks or stress, can also contribute to heart failure or a heart attack. While these events cannot be predicted, continuous data from patients’ health can help doctors predict heart failure. Therefore, this data-driven research utilizes advanced machine learning and deep learning techniques to better analyze and manipulate the data, providing doctors with informative decision-making tools regarding a person’s likelihood of experiencing heart failure. In this paper, the author employed advanced data preprocessing and cleaning techniques. Additionally, the dataset underwent testing using two different methodologies to determine the most effective machine-learning technique for producing optimal predictions. The first methodology involved employing a list of supervised classification machine learning algorithms, including Naïve Bayes (NB), KNN, logistic regression, and the SVM algorithm. The second methodology utilized a deep learning (DL) algorithm known as Multilayer Perceptrons (MLPs). This algorithm provided the author with the flexibility to experiment with different layer sizes and activation functions, such as ReLU, logistic (sigmoid), and Tanh. Both methodologies produced optimal models with high-level accuracy rates. The first methodology involves a list of supervised machine learning algorithms, including KNN, SVM, Adaboost, Logistic Regression, Naive Bayes, and Decision Tree algorithms. They achieved accuracy rates of 86%, 89%, 89%, 81%, 79%, and 99%, respectively. The author clearly explained that Decision Tree algorithm is not suitable for the dataset at hand due to overfitting issues. Therefore, it was discarded as an optimal model to be used. However, the latter methodology (Neural Network) demonstrated the most stable and optimal accuracy, achieving over 87% accuracy while adapting well to real-life situations and requiring low computing power overall. A performance assessment and evaluation were carried out based on a confusion matrix report to demonstrate feasibility and performance. The author concluded that the performance of the model in real-life situations can advance not only the medical field of science but also mathematical concepts. Additionally, the advanced preprocessing approach behind the model can provide value to the Data Science community. The model can be further developed by employing various optimization techniques to handle even larger datasets related to heart failures. Furthermore, different neural network algorithms can be tested to explore alternative approaches and yield different results.展开更多
The development of accurate prediction models continues to be highly beneficial in myriad disciplines. Deep learning models have performed well in stock price prediction and give high accuracy. However, these models a...The development of accurate prediction models continues to be highly beneficial in myriad disciplines. Deep learning models have performed well in stock price prediction and give high accuracy. However, these models are largely affected by the vanishing gradient problem escalated by some activation functions. This study proposes the use of the Vanishing Gradient Resilient Optimized Gated Recurrent Unit (OGRU) model with a scaled mean Approximation Coefficient (AC) time lag which should counter slow convergence, vanishing gradient and large error metrics. This study employed the Rectified Linear Unit (ReLU), Hyperbolic Tangent (Tanh), Sigmoid and Exponential Linear Unit (ELU) activation functions. Real-life datasets including the daily Apple and 5-minute Netflix closing stock prices were used, and they were decomposed using the Stationary Wavelet Transform (SWT). The decomposed series formed a decomposed data model which was compared to an undecomposed data model with similar hyperparameters and different default lags. The Apple daily dataset performed well with a Default_1 lag, using an undecomposed data model and the ReLU, attaining 0.01312, 0.00854 and 3.67 minutes for RMSE, MAE and runtime. The Netflix data performed best with the MeanAC_42 lag, using decomposed data model and the ELU achieving 0.00620, 0.00487 and 3.01 minutes for the same metrics.展开更多
Actinomycetes are relatively prevalent bacteria in the ocean,constituting 9% of the total number of marine bacteria.The advancement of science and technology has led to a more profound exploration of marine actinomyce...Actinomycetes are relatively prevalent bacteria in the ocean,constituting 9% of the total number of marine bacteria.The advancement of science and technology has led to a more profound exploration of marine actinomycetes.These studies hold immense significance in comprehending the distribution and adaptation of marine actinomycetes within the oceanic environment,as well as uncovering new secondary metabolites.Based on differing lifestyles,marine actinomycetes can be categorized as free-living or co-epiphytic.The activity and metabolism of actinomycetes vary across diverse marine settings,including the deep sea,benthic regions,and marine organisms.Due to their distinctive biological traits and genetic background,these marine actinomycetes inevitably generate metabolites possessing unique structures.Research methodologies concerning marine actinomycetes predominantly encompass traditional pure culture techniques,molecular biology approaches,and the integration of metagenomics and bioinformatics.The exploration of varied methodologies proves pivotal for the analysis of metabolite processes.Through the cultivation of marine actinomycetes,numerous compounds featuring novel structures and significant activities have been isolated,furnishing a substantial foundation for new drug investigations.These encompass,but are not restricted to,peptides,antibiotics,terpenoids,ketones,quinones,macrolides,and pigments.The potential applications of marine actinomyces and their secondary metabolites extend beyond antibacterial and anti-tumor effects,exhibiting promising prospects in antifungal and antiviral domains.This paper provides a comprehensive review of the classification,resources,research methodologies,and habitats of marine actinomycetes.Furthermore,it delves into the classification of secondary metabolites and their functional activities,facilitating a more exhaustive analysis of the secondary metabolites produced by marine actinomycetes.展开更多
文摘Activation functions play an essential role in converting the output of the artificial neural network into nonlinear results,since without this nonlinearity,the results of the network will be less accurate.Nonlinearity is the mission of all nonlinear functions,except for polynomials.The activation function must be dif-ferentiable for backpropagation learning.This study’s objective is to determine the best activation functions for the approximation of each fractal image.Different results have been attained using Matlab and Visual Basic programs,which indi-cate that the bounded function is more helpful than other functions.The non-lin-earity of the activation function is important when using neural networks for coding fractal images because the coefficients of the Iterated Function System are different according to the different types of fractals.The most commonly cho-sen activation function is the sigmoidal function,which produces a positive value.Other functions,such as tansh or arctan,whose values can be positive or negative depending on the network input,tend to train neural networks faster.The coding speed of the fractal image is different depending on the appropriate activation function chosen for each fractal shape.In this paper,we have provided the appro-priate activation functions for each type of system of iterated functions that help the network to identify the transactions of the system.
基金The authors greatly thanked the financial support from the National Key Research and Development Program of China(funded by National Natural Science Foundation of China,No.2019YFA0708300)the Strategic Cooperation Technology Projects of CNPC and CUPB(funded by China National Petroleum Corporation,No.ZLZX2020-03)+1 种基金the National Science Fund for Distinguished Young Scholars(funded by National Natural Science Foundation of China,No.52125401)Science Foundation of China University of Petroleum,Beijing(funded by China University of petroleum,Beijing,No.2462022SZBH002).
文摘Accurate prediction of the rate of penetration(ROP)is significant for drilling optimization.While the intelligent ROP prediction model based on fully connected neural networks(FNN)outperforms traditional ROP equations and machine learning algorithms,its lack of interpretability undermines its credibility.This study proposes a novel interpretation and characterization method for the FNN ROP prediction model using the Rectified Linear Unit(ReLU)activation function.By leveraging the derivative of the ReLU function,the FNN function calculation process is transformed into vector operations.The FNN model is linearly characterized through further simplification,enabling its interpretation and analysis.The proposed method is applied in ROP prediction scenarios using drilling data from three vertical wells in the Tarim Oilfield.The results demonstrate that the FNN ROP prediction model with ReLU as the activation function performs exceptionally well.The relative activation frequency curve of hidden layer neurons aids in analyzing the overfitting of the FNN ROP model and determining drilling data similarity.In the well sections with similar drilling data,averaging the weight parameters enables linear characterization of the FNN ROP prediction model,leading to the establishment of a corresponding linear representation equation.Furthermore,the quantitative analysis of each feature's influence on ROP facilitates the proposal of drilling parameter optimization schemes for the current well section.The established linear characterization equation exhibits high precision,strong stability,and adaptability through the application and validation across multiple well sections.
基金funded by the National Natural Science Foundation of China(NSFC)(No.52274222)research project supported by Shanxi Scholarship Council of China(No.2023-036).
文摘This paper presents a novel framework aimed at quantifying uncertainties associated with the 3D reconstruction of smoke from2Dimages.This approach reconstructs color and density fields from 2D images using Neural Radiance Field(NeRF)and improves image quality using frequency regularization.The NeRF model is obtained via joint training ofmultiple artificial neural networks,whereby the expectation and standard deviation of density fields and RGB values can be evaluated for each pixel.In addition,customized physics-informed neural network(PINN)with residual blocks and two-layer activation functions are utilized to input the density fields of the NeRF into Navier-Stokes equations and convection-diffusion equations to reconstruct the velocity field.The velocity uncertainties are also evaluated through ensemble learning.The effectiveness of the proposed algorithm is demonstrated through numerical examples.The presentmethod is an important step towards downstream tasks such as reliability analysis and robust optimization in engineering design.
文摘The development of defect prediction plays a significant role in improving software quality. Such predictions are used to identify defective modules before the testing and to minimize the time and cost. The software with defects negatively impacts operational costs and finally affects customer satisfaction. Numerous approaches exist to predict software defects. However, the timely and accurate software bugs are the major challenging issues. To improve the timely and accurate software defect prediction, a novel technique called Nonparametric Statistical feature scaled QuAdratic regressive convolution Deep nEural Network (SQADEN) is introduced. The proposed SQADEN technique mainly includes two major processes namely metric or feature selection and classification. First, the SQADEN uses the nonparametric statistical Torgerson–Gower scaling technique for identifying the relevant software metrics by measuring the similarity using the dice coefficient. The feature selection process is used to minimize the time complexity of software fault prediction. With the selected metrics, software fault perdition with the help of the Quadratic Censored regressive convolution deep neural network-based classification. The deep learning classifier analyzes the training and testing samples using the contingency correlation coefficient. The softstep activation function is used to provide the final fault prediction results. To minimize the error, the Nelder–Mead method is applied to solve non-linear least-squares problems. Finally, accurate classification results with a minimum error are obtained at the output layer. Experimental evaluation is carried out with different quantitative metrics such as accuracy, precision, recall, F-measure, and time complexity. The analyzed results demonstrate the superior performance of our proposed SQADEN technique with maximum accuracy, sensitivity and specificity by 3%, 3%, 2% and 3% and minimum time and space by 13% and 15% when compared with the two state-of-the-art methods.
基金supported by Defence Innovative Research Program(DIRP)Grant(PA No.9015102335)from Defence Research&Technology Office,Ministry of Defence,Singapore。
文摘Background:Excessive heat exposure can lead to hyperthermia in humans,which impairs physical performance and disrupts cognitive function.While heat is a known physiological stressor,it is unclear how severe heat stress affects brain physiology and function.Methods:Eleven healthy participants were subjected to heat stress from prolonged exercise or warm water immersion until their rectal temperatures(T_(re))attained 39.5℃,inducing exertional or passive hyperthermia,respectively.In a separate trial,blended ice was ingested before and during exercise as a cooling strategy.Data were compared to a control condition with seated rest(normothermic).Brain temperature(T_(br)),cerebral perfusion,and task-based brain activity were assessed using magnetic resonance imaging techniques.Results:T_(br)in motor cortex was found to be tightly regulated at rest(37.3℃±0.4℃(mean±SD))despite fluctuations in T_(re).With the development of hyperthermia,T_(br)increases and dovetails with the rising T_(re).Bilateral motor cortical activity was suppressed during high-intensity plantarflexion tasks,implying a reduced central motor drive in hyperthermic participants(T_(re)=38.5℃±0.1℃).Global gray matter perfusion and regional perfusion in sensorimotor cortex were reduced with passive hyperthermia.Executive function was poorer under a passive hyperthermic state,and this could relate to compromised visual processing as indicated by the reduced activation of left lateral-occipital cortex.Conversely,ingestion of blended ice before and during exercise alleviated the rise in both T_(re)and T_(bc)and mitigated heat-related neural perturbations.Conclusion:Severe heat exposure elevates T_(br),disrupts motor cortical activity and executive function,and this can lead to impairment of physical and cognitive performance.
基金National Key R&D Program of China(2018AAA0102600)National Natural Science Foundation of China(No.61876215,62106119)+1 种基金Beijing Academy of Artificial Intelligence(BAAI),ChinaChinese Institute for Brain Research,Beijing,and the Science and Technology Major Project of Guangzhou,China(202007030006).
文摘Self-normalizing neural networks(SNN)regulate the activation and gradient flows through activation functions with the self-normalization property.As SNNs do not rely on norms computed from minibatches,they are more friendly to data parallelism,kernel fusion,and emerging architectures such as ReRAM-based accelerators.However,existing SNNs have mainly demonstrated their effectiveness on toy datasets and fall short in accuracy when dealing with large-scale tasks like ImageNet.They lack the strong normalization,regularization,and expression power required for wider,deeper models and larger-scale tasks.To enhance the normalization strength,this paper introduces a comprehensive and practical definition of the self-normalization property in terms of the stability and attractiveness of the statistical fixed points.It is comprehensive as it jointly considers all the fixed points used by existing studies:the first and second moment of forward activation and the expected Frobenius norm of backward gradient.The practicality comes from the analytical equations provided by our paper to assess the stability and attractiveness of each fixed point,which are derived from theoretical analysis of the forward and backward signals.The proposed definition is applied to a meta activation function inspired by prior research,leading to a stronger self-normalizing activation function named‘‘bi-scaled exponential linear unit with backward standardized’’(bSELU-BSTD).We provide both theoretical and empirical evidence to show that it is superior to existing studies.To enhance the regularization and expression power,we further propose scaled-Mixup and channel-wise scale&shift.With these three techniques,our approach achieves 75.23%top-1 accuracy on the ImageNet with Conv MobileNet V1,surpassing the performance of existing self-normalizing activation functions.To the best of our knowledge,this is the first SNN that achieves comparable accuracy to batch normalization on ImageNet.
文摘Recently we have studied the rare earth ion-selective electrodes with active materials of the func-tional polymers and found that the process chosen for the functional polymers had an effect on the propertiesof gadolinium ion selective electrode besides the effects of their structures.1.Effect of preparation process of the grafted polymers on the properties ofgadolinium ion selective electrodesThe electrode membranes which consist of functional polymers as active materials were prepared by re-action of gadolinium chloride with the radiation grafted clmer of acrlic acid and polystyrene of which
文摘This paper describes our implementation of several neural networks built on a field programmable gate array (FPGA) and used to recognize a handwritten digit dataset—the Modified National Institute of Standards and Technology (MNIST) database. We also propose a novel hardware-friendly activation function called the dynamic Rectifid Linear Unit (ReLU)—D-ReLU function that achieves higher performance than traditional activation functions at no cost to accuracy. We built a 2-layer online training multilayer perceptron (MLP) neural network on an FPGA with varying data width. Reducing the data width from 8 to 4 bits only reduces prediction accuracy by 11%, but the FPGA area decreases by 41%. Compared to networks that use the sigmoid functions, our proposed D-ReLU function uses 24% - 41% less area with no loss to prediction accuracy. Further reducing the data width of the 3-layer networks from 8 to 4 bits, the prediction accuracies only decrease by 3% - 5%, with area being reduced by 9% - 28%. Moreover, FPGA solutions have 29 times faster execution time, even despite running at a 60× lower clock rate. Thus, FPGA implementations of neural networks offer a high-performance, low power alternative to traditional software methods, and our novel D-ReLU activation function offers additional improvements to performance and power saving.
基金Project supported by the National Natural Science Foundation of China(Grant Nos.61374094 and 61503338)the Natural Science Foundation of Zhejiang Province,China(Grant No.LQ15F030005)
文摘In this paper, the multistability issue is discussed for delayed complex-valued recurrent neural networks with discontinuous real-imaginary-type activation functions. Based on a fixed theorem and stability definition, sufficient criteria are established for the existence and stability of multiple equilibria of complex-valued recurrent neural networks. The number of stable equilibria is larger than that of real-valued recurrent neural networks, which can be used to achieve high-capacity associative memories. One numerical example is provided to show the effectiveness and superiority of the presented results.
文摘In this paper,the functional polymeric active materials were prepared by the grafting copolymerization and their structure and properties were studied.The results show that the structure and properties of these ac- tive materials have the relative large effects on the properties of gadolinium ion selective electrodes.
文摘Respect for human rights and protection of human rights are significant rules in the Constitution of the People’s Republic of China.In 2015,judicial administration departments at all levels legally exercised their duties,implemented the principles and rules in constitution,and kept strengthening propagation of human rights through creation of contents and methods which had acquired great effects.What they have done contributes significantly toward the development of human rights in China.
基金supported by the National Natural Science Foundation of China (Grant No.62103432)supported by Young Talent fund of University Association for Science and Technology in Shaanxi, China(Grant No.20210108)。
文摘An impact point prediction(IPP) guidance based on supervised learning is proposed to address the problem of precise guidance for the ballistic missile in high maneuver penetration condition.An accurate ballistic trajectory model is applied to generate training samples,and ablation experiments are conducted to determine the mapping relationship between the flight state and the impact point.At the same time,the impact point coordinates are decoupled to improve the prediction accuracy,and the sigmoid activation function is improved to ameliorate the prediction efficiency.Therefore,an IPP neural network model,which solves the contradiction between the accuracy and the speed of the IPP,is established.In view of the performance deviation of the divert control system,the mapping relationship between the guidance parameters and the impact deviation is analysed based on the variational principle.In addition,a fast iterative model of guidance parameters is designed for reference to the Newton iteration method,which solves the nonlinear strong coupling problem of the guidance parameter solution.Monte Carlo simulation results show that the prediction accuracy of the impact point is high,with a 3 σ prediction error of 4.5 m,and the guidance method is robust,with a 3 σ error of 7.5 m.On the STM32F407 singlechip microcomputer,a single IPP takes about 2.374 ms,and a single guidance solution takes about9.936 ms,which has a good real-time performance and a certain engineering application value.
基金This work was supported by the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(No.2022R1F1A1062953).
文摘Recently,deep learning has achieved remarkable results in fields that require human cognitive ability,learning ability,and reasoning ability.Activation functions are very important because they provide the ability of artificial neural networks to learn complex patterns through nonlinearity.Various activation functions are being studied to solve problems such as vanishing gradients and dying nodes that may occur in the deep learning process.However,it takes a lot of time and effort for researchers to use the existing activation function in their research.Therefore,in this paper,we propose a universal activation function(UA)so that researchers can easily create and apply various activation functions and improve the performance of neural networks.UA can generate new types of activation functions as well as functions like traditional activation functions by properly adjusting three hyperparameters.The famous Convolutional Neural Network(CNN)and benchmark datasetwere used to evaluate the experimental performance of the UA proposed in this study.We compared the performance of the artificial neural network to which the traditional activation function is applied and the artificial neural network to which theUA is applied.In addition,we evaluated the performance of the new activation function generated by adjusting the hyperparameters of theUA.The experimental performance evaluation results showed that the classification performance of CNNs improved by up to 5%through the UA,although most of them showed similar performance to the traditional activation function.
文摘Machine Learning(ML)and Deep Learning(DL)technologies are revolutionizing the medical domain,especially with Electrocardiogram(ECG),by providing new tools and techniques for diagnosing,treating,and preventing diseases.However,DL architectures are computationally more demanding.In recent years,researchers have focused on combining the computationally less intensive portion of the DL architectures with ML approaches,say for example,combining the convolutional layer blocks of Convolution Neural Networks(CNNs)into ML algorithms such as Extreme Gradient Boosting(XGBoost)and K-Nearest Neighbor(KNN)resulting in CNN-XGBoost and CNN-KNN,respectively.However,these approaches are homogenous in the sense that they use a fixed Activation Function(AFs)in the sequence of convolution and pooling layers,thereby limiting the ability to capture unique features.Since various AFs are readily available and each could capture unique features,we propose a Convolutionbased Heterogeneous Activation Facility(CHAF)which uses multiple AFs in the convolution layer blocks,one for each block,with a motivation of extracting features in a better manner to improve the accuracy.The proposed CHAF approach is validated on PTB and shown to outperform the homogeneous approaches such as CNN-KNN and CNN-XGBoost.For PTB dataset,proposed CHAF-KNN has an accuracy of 99.55%and an F1 score of 99.68%in just 0.008 s,outperforming the state-of-the-art CNN-XGBoost which has an accuracy of 99.38%and an F1 score of 99.32%in 1.23 s.To validate the generality of the proposed CHAF,experiments were repeated on MIT-BIH dataset,and the proposed CHAF-KNN is shown to outperform CNN-KNN and CNN-XGBoost.
文摘Renewable energy production plays a major role in satisfying electricity demand.Wind power conversion is one of the most popular renewable energy sources compared to other sources.Wind energy conversion has two major types of generators such as the Permanent Magnet Synchronous Generator(PMSG)and the Doubly Fed Induction Generator(DFIG).The maximum power tracking algo-rithm is a crucial controller,a wind energy conversion system for generating maximum power in different wind speed conditions.In this article,the DFIG wind energy conversion system was developed in Matrix Laboratory(MATLAB)and designed a machine learning(ML)algorithm for the rotor and grid side converter.The ML algorithm has been developed and trained in a MATLAB environment.There are two types of learning algorithms such as supervised and unsupervised learning.In this research supervised learning is used to power the neural networks and analysis is made for various hidden layers and activation functions.Simulation results are assessed to demonstrate the efficiency of the proposed system.
文摘Multi-port converters are considered as exceeding earlier period decade owing to function in a combination of different energy sources in a single processing unit.Renewable energy sources are playing a significant role in the modern energy system with rapid development.In renewable sources like fuel combustion and solar energy,the generated voltages change due to their environmental changes.To develop energy resources,electric power generation involved huge awareness.The power and output voltages are plays important role in our work but it not considered in the existing system.For considering the power and voltage,Gaussian PI Controller-Maxpooling Deep Convolutional Neural Network Classifier(GPIC-MDCNNC)Model is introduced for the grid-connected renewable energy system.The input information is collected from two input sources.After that,input layer transfer information to hidden layer 1 fuzzy PI is employed for controlling voltage in GPIC-MDCNNC Model.Hidden layer 1 is transferred to hidden layer 2.Gaussian activation is employed for determining the output voltage with help of the controller.At last,the output layer offers the last value in GPIC-MDCNNC Model.The designed method was confirmed using one and multiple sources by stable and unpredictable input voltages.GPIC-MDCNNC Model increases the performance of grid-connected renewable energy systems by enhanced voltage value compared with state-of-the-art works.The control technique using GPIC-MDCNNC Model increases the dynamics of hybrid energy systems connected to the grid.
基金Supported by Science and Technology Research Project of Hebei Provincial Department of Education (ZC2021206)。
文摘Ginkgo biloba resources in China are enormous. With the demand of the market, the preparation and application of flavonoids have become a current research hotspot. The main active substances in G. biloba leaves, flavonoids, have various functional activities and are widely used in fields such as food, medicine, cosmetics, feed, etc. In this paper, the introduction, functional activity, extraction methods, and application research of flavonoids from G. biloba leaves are reviewed, and the development prospects of flavonoids from G. biloba leaves are expected.
文摘In this study, the author will investigate and utilize advanced machine learning models related to two different methodologies to determine the best and most effective way to predict individuals with heart failure and cardiovascular diseases. The first methodology involves a list of classification machine learning algorithms, and the second methodology involves the use of a deep learning algorithm known as MLP or Multilayer Perceptrons. Globally, hospitals are dealing with cases related to cardiovascular diseases and heart failure as they are major causes of death, not only for overweight individuals but also for those who do not adopt a healthy diet and lifestyle. Often, heart failures and cardiovascular diseases can be caused by many factors, including cardiomyopathy, high blood pressure, coronary heart disease, and heart inflammation [1]. Other factors, such as irregular shocks or stress, can also contribute to heart failure or a heart attack. While these events cannot be predicted, continuous data from patients’ health can help doctors predict heart failure. Therefore, this data-driven research utilizes advanced machine learning and deep learning techniques to better analyze and manipulate the data, providing doctors with informative decision-making tools regarding a person’s likelihood of experiencing heart failure. In this paper, the author employed advanced data preprocessing and cleaning techniques. Additionally, the dataset underwent testing using two different methodologies to determine the most effective machine-learning technique for producing optimal predictions. The first methodology involved employing a list of supervised classification machine learning algorithms, including Naïve Bayes (NB), KNN, logistic regression, and the SVM algorithm. The second methodology utilized a deep learning (DL) algorithm known as Multilayer Perceptrons (MLPs). This algorithm provided the author with the flexibility to experiment with different layer sizes and activation functions, such as ReLU, logistic (sigmoid), and Tanh. Both methodologies produced optimal models with high-level accuracy rates. The first methodology involves a list of supervised machine learning algorithms, including KNN, SVM, Adaboost, Logistic Regression, Naive Bayes, and Decision Tree algorithms. They achieved accuracy rates of 86%, 89%, 89%, 81%, 79%, and 99%, respectively. The author clearly explained that Decision Tree algorithm is not suitable for the dataset at hand due to overfitting issues. Therefore, it was discarded as an optimal model to be used. However, the latter methodology (Neural Network) demonstrated the most stable and optimal accuracy, achieving over 87% accuracy while adapting well to real-life situations and requiring low computing power overall. A performance assessment and evaluation were carried out based on a confusion matrix report to demonstrate feasibility and performance. The author concluded that the performance of the model in real-life situations can advance not only the medical field of science but also mathematical concepts. Additionally, the advanced preprocessing approach behind the model can provide value to the Data Science community. The model can be further developed by employing various optimization techniques to handle even larger datasets related to heart failures. Furthermore, different neural network algorithms can be tested to explore alternative approaches and yield different results.
文摘The development of accurate prediction models continues to be highly beneficial in myriad disciplines. Deep learning models have performed well in stock price prediction and give high accuracy. However, these models are largely affected by the vanishing gradient problem escalated by some activation functions. This study proposes the use of the Vanishing Gradient Resilient Optimized Gated Recurrent Unit (OGRU) model with a scaled mean Approximation Coefficient (AC) time lag which should counter slow convergence, vanishing gradient and large error metrics. This study employed the Rectified Linear Unit (ReLU), Hyperbolic Tangent (Tanh), Sigmoid and Exponential Linear Unit (ELU) activation functions. Real-life datasets including the daily Apple and 5-minute Netflix closing stock prices were used, and they were decomposed using the Stationary Wavelet Transform (SWT). The decomposed series formed a decomposed data model which was compared to an undecomposed data model with similar hyperparameters and different default lags. The Apple daily dataset performed well with a Default_1 lag, using an undecomposed data model and the ReLU, attaining 0.01312, 0.00854 and 3.67 minutes for RMSE, MAE and runtime. The Netflix data performed best with the MeanAC_42 lag, using decomposed data model and the ELU achieving 0.00620, 0.00487 and 3.01 minutes for the same metrics.
文摘Actinomycetes are relatively prevalent bacteria in the ocean,constituting 9% of the total number of marine bacteria.The advancement of science and technology has led to a more profound exploration of marine actinomycetes.These studies hold immense significance in comprehending the distribution and adaptation of marine actinomycetes within the oceanic environment,as well as uncovering new secondary metabolites.Based on differing lifestyles,marine actinomycetes can be categorized as free-living or co-epiphytic.The activity and metabolism of actinomycetes vary across diverse marine settings,including the deep sea,benthic regions,and marine organisms.Due to their distinctive biological traits and genetic background,these marine actinomycetes inevitably generate metabolites possessing unique structures.Research methodologies concerning marine actinomycetes predominantly encompass traditional pure culture techniques,molecular biology approaches,and the integration of metagenomics and bioinformatics.The exploration of varied methodologies proves pivotal for the analysis of metabolite processes.Through the cultivation of marine actinomycetes,numerous compounds featuring novel structures and significant activities have been isolated,furnishing a substantial foundation for new drug investigations.These encompass,but are not restricted to,peptides,antibiotics,terpenoids,ketones,quinones,macrolides,and pigments.The potential applications of marine actinomyces and their secondary metabolites extend beyond antibacterial and anti-tumor effects,exhibiting promising prospects in antifungal and antiviral domains.This paper provides a comprehensive review of the classification,resources,research methodologies,and habitats of marine actinomycetes.Furthermore,it delves into the classification of secondary metabolites and their functional activities,facilitating a more exhaustive analysis of the secondary metabolites produced by marine actinomycetes.