Most of the neural network architectures are based on human experience,which requires a long and tedious trial-and-error process.Neural architecture search(NAS)attempts to detect effective architectures without human ...Most of the neural network architectures are based on human experience,which requires a long and tedious trial-and-error process.Neural architecture search(NAS)attempts to detect effective architectures without human intervention.Evolutionary algorithms(EAs)for NAS can find better solutions than human-designed architectures by exploring a large search space for possible architectures.Using multiobjective EAs for NAS,optimal neural architectures that meet various performance criteria can be explored and discovered efficiently.Furthermore,hardware-accelerated NAS methods can improve the efficiency of the NAS.While existing reviews have mainly focused on different strategies to complete NAS,a few studies have explored the use of EAs for NAS.In this paper,we summarize and explore the use of EAs for NAS,as well as large-scale multiobjective optimization strategies and hardware-accelerated NAS methods.NAS performs well in healthcare applications,such as medical image analysis,classification of disease diagnosis,and health monitoring.EAs for NAS can automate the search process and optimize multiple objectives simultaneously in a given healthcare task.Deep neural network has been successfully used in healthcare,but it lacks interpretability.Medical data is highly sensitive,and privacy leaks are frequently reported in the healthcare industry.To solve these problems,in healthcare,we propose an interpretable neuroevolution framework based on federated learning to address search efficiency and privacy protection.Moreover,we also point out future research directions for evolutionary NAS.Overall,for researchers who want to use EAs to optimize NNs in healthcare,we analyze the advantages and disadvantages of doing so to provide detailed guidance,and propose an interpretable privacy-preserving framework for healthcare applications.展开更多
Deep neural networks often outperform classical machine learning algorithms in solving real-world problems.However,designing better networks usually requires domain expertise and consumes significant time and com-puti...Deep neural networks often outperform classical machine learning algorithms in solving real-world problems.However,designing better networks usually requires domain expertise and consumes significant time and com-puting resources.Moreover,when the task changes,the original network architecture becomes outdated and requires redesigning.Thus,Neural Architecture Search(NAS)has gained attention as an effective approach to automatically generate optimal network architectures.Most NAS methods mainly focus on achieving high performance while ignoring architectural complexity.A myriad of research has revealed that network performance and structural complexity are often positively correlated.Nevertheless,complex network structures will bring enormous computing resources.To cope with this,we formulate the neural architecture search task as a multi-objective optimization problem,where an optimal architecture is learned by minimizing the classification error rate and the number of network parameters simultaneously.And then a decomposition-based multi-objective stochastic fractal search method is proposed to solve it.In view of the discrete property of the NAS problem,we discretize the stochastic fractal search step size so that the network architecture can be optimized more effectively.Additionally,two distinct update methods are employed in step size update stage to enhance the global and local search abilities adaptively.Furthermore,an information exchange mechanism between architectures is raised to accelerate the convergence process and improve the efficiency of the algorithm.Experimental studies show that the proposed algorithm has competitive performance comparable to many existing manual and automatic deep neural network generation approaches,which achieved a parameter-less and high-precision architecture with low-cost on each of the six benchmark datasets.展开更多
In this paper,a novel method of ultra-lightweight convolution neural network(CNN)design based on neural architecture search(NAS)and knowledge distillation(KD)is proposed.It can realize the automatic construction of th...In this paper,a novel method of ultra-lightweight convolution neural network(CNN)design based on neural architecture search(NAS)and knowledge distillation(KD)is proposed.It can realize the automatic construction of the space target inverse synthetic aperture radar(ISAR)image recognition model with ultra-lightweight and high accuracy.This method introduces the NAS method into the radar image recognition for the first time,which solves the time-consuming and labor-consuming problems in the artificial design of the space target ISAR image automatic recognition model(STIIARM).On this basis,the NAS model’s knowledge is transferred to the student model with lower computational complexity by the flow of the solution procedure(FSP)distillation method.Thus,the decline of recognition accuracy caused by the direct compression of model structural parameters can be effectively avoided,and the ultralightweight STIIARM can be obtained.In the method,the Inverted Linear Bottleneck(ILB)and Inverted Residual Block(IRB)are firstly taken as each block’s basic structure in CNN.And the expansion ratio,output filter size,number of IRBs,and convolution kernel size are set as the search parameters to construct a hierarchical decomposition search space.Then,the recognition accuracy and computational complexity are taken as the objective function and constraint conditions,respectively,and the global optimization model of the CNN architecture search is established.Next,the simulated annealing(SA)algorithm is used as the search strategy to search out the lightweight and high accuracy STIIARM directly.After that,based on the three principles of similar block structure,the same corresponding channel number,and the minimum computational complexity,the more lightweight student model is designed,and the FSP matrix pairing between the NAS model and student model is completed.Finally,by minimizing the loss between the FSP matrix pairs of the NAS model and student model,the student model’s weight adjustment is completed.Thus the ultra-lightweight and high accuracy STIIARM is obtained.The proposed method’s effectiveness is verified by the simulation experiments on the ISAR image dataset of five types of space targets.展开更多
Neural architecture search(NAS)has become increasingly popular in the deep learning community recently,mainly because it can provide an opportunity to allow interested users without rich expertise to benefit from the ...Neural architecture search(NAS)has become increasingly popular in the deep learning community recently,mainly because it can provide an opportunity to allow interested users without rich expertise to benefit from the success of deep neural networks(DNNs).However,NAS is still laborious and time-consuming because a large number of performance estimations are required during the search process of NAS,and training DNNs is computationally intensive.To solve this major limitation of NAS,improving the computational efficiency is essential in the design of NAS.However,a systematic overview of computationally efficient NAS(CE-NAS)methods still lacks.To fill this gap,we provide a comprehensive survey of the state-of-the-art on CE-NAS by categorizing the existing work into proxy-based and surrogate-assisted NAS methods,together with a thorough discussion of their design principles and a quantitative comparison of their performances and computational complexities.The remaining challenges and open research questions are also discussed,and promising research topics in this emerging field are suggested.展开更多
AutoML(Automated Machine Learning)is an emerging field that aims to automate the process of building machine learning models.AutoML emerged to increase productivity and efficiency by automating as much as possible the...AutoML(Automated Machine Learning)is an emerging field that aims to automate the process of building machine learning models.AutoML emerged to increase productivity and efficiency by automating as much as possible the inefficient work that occurs while repeating this process whenever machine learning is applied.In particular,research has been conducted for a long time on technologies that can effectively develop high-quality models by minimizing the intervention of model developers in the process from data preprocessing to algorithm selection and tuning.In this semantic review research,we summarize the data processing requirements for AutoML approaches and provide a detailed explanation.We place greater emphasis on neural architecture search(NAS)as it currently represents a highly popular sub-topic within the field of AutoML.NAS methods use machine learning algorithms to search through a large space of possible architectures and find the one that performs best on a given task.We provide a summary of the performance achieved by representative NAS algorithms on the CIFAR-10,CIFAR-100,ImageNet and wellknown benchmark datasets.Additionally,we delve into several noteworthy research directions in NAS methods including one/two-stage NAS,one-shot NAS and joint hyperparameter with architecture optimization.We discussed how the search space size and complexity in NAS can vary depending on the specific problem being addressed.To conclude,we examine several open problems(SOTA problems)within current AutoML methods that assure further investigation in future research.展开更多
Deep learning has emerged in many practical applications,such as image classification,fault diagnosis,and object detection.More recently,convolutional neural networks(CNNs),representative models of deep learning,have ...Deep learning has emerged in many practical applications,such as image classification,fault diagnosis,and object detection.More recently,convolutional neural networks(CNNs),representative models of deep learning,have been used to solve fault detection.However,the current design of CNNs for fault detection of wind turbine blades is highly dependent on domain knowledge and requires a large amount of trial and error.For this reason,an evolutionary YOLOv8 network has been developed to automatically find the network architecture for wind turbine blade-based fault detection.YOLOv8 is a CNN-backed object detection model.Specifically,to reduce the parameter count,we first design an improved FasterNet module based on the Partial Convolution(PConv)operator.Then,to enhance convergence performance,we improve the loss function based on the efficient complete intersection over the union.Based on this,a flexible variable-length encoding is proposed,and the corresponding reproduction operators are designed.Related experimental results confirmthat the proposed approach can achieve better fault detection results and improve by 2.6%in mean precision at 50(mAP50)compared to the existing methods.Additionally,compared to training with the YOLOv8n model,the YOLOBFE model reduces the training parameters by 933,937 and decreases the GFLOPS(Giga Floating Point Operations Per Second)by 1.1.展开更多
基金supported in part by the National Natural Science Foundation of China (NSFC) under Grant No.61976242in part by the Natural Science Fund of Hebei Province for Distinguished Young Scholars under Grant No.F2021202010+2 种基金in part by the Fundamental Scientific Research Funds for Interdisciplinary Team of Hebei University of Technology under Grant No.JBKYTD2002funded by Science and Technology Project of Hebei Education Department under Grant No.JZX2023007supported by 2022 Interdisciplinary Postgraduate Training Program of Hebei University of Technology under Grant No.HEBUT-YXKJC-2022122.
文摘Most of the neural network architectures are based on human experience,which requires a long and tedious trial-and-error process.Neural architecture search(NAS)attempts to detect effective architectures without human intervention.Evolutionary algorithms(EAs)for NAS can find better solutions than human-designed architectures by exploring a large search space for possible architectures.Using multiobjective EAs for NAS,optimal neural architectures that meet various performance criteria can be explored and discovered efficiently.Furthermore,hardware-accelerated NAS methods can improve the efficiency of the NAS.While existing reviews have mainly focused on different strategies to complete NAS,a few studies have explored the use of EAs for NAS.In this paper,we summarize and explore the use of EAs for NAS,as well as large-scale multiobjective optimization strategies and hardware-accelerated NAS methods.NAS performs well in healthcare applications,such as medical image analysis,classification of disease diagnosis,and health monitoring.EAs for NAS can automate the search process and optimize multiple objectives simultaneously in a given healthcare task.Deep neural network has been successfully used in healthcare,but it lacks interpretability.Medical data is highly sensitive,and privacy leaks are frequently reported in the healthcare industry.To solve these problems,in healthcare,we propose an interpretable neuroevolution framework based on federated learning to address search efficiency and privacy protection.Moreover,we also point out future research directions for evolutionary NAS.Overall,for researchers who want to use EAs to optimize NNs in healthcare,we analyze the advantages and disadvantages of doing so to provide detailed guidance,and propose an interpretable privacy-preserving framework for healthcare applications.
基金supported by the China Postdoctoral Science Foundation Funded Project(Grant Nos.2017M613054 and 2017M613053)the Shaanxi Postdoctoral Science Foundation Funded Project(Grant No.2017BSHYDZZ33)the National Science Foundation of China(Grant No.62102239).
文摘Deep neural networks often outperform classical machine learning algorithms in solving real-world problems.However,designing better networks usually requires domain expertise and consumes significant time and com-puting resources.Moreover,when the task changes,the original network architecture becomes outdated and requires redesigning.Thus,Neural Architecture Search(NAS)has gained attention as an effective approach to automatically generate optimal network architectures.Most NAS methods mainly focus on achieving high performance while ignoring architectural complexity.A myriad of research has revealed that network performance and structural complexity are often positively correlated.Nevertheless,complex network structures will bring enormous computing resources.To cope with this,we formulate the neural architecture search task as a multi-objective optimization problem,where an optimal architecture is learned by minimizing the classification error rate and the number of network parameters simultaneously.And then a decomposition-based multi-objective stochastic fractal search method is proposed to solve it.In view of the discrete property of the NAS problem,we discretize the stochastic fractal search step size so that the network architecture can be optimized more effectively.Additionally,two distinct update methods are employed in step size update stage to enhance the global and local search abilities adaptively.Furthermore,an information exchange mechanism between architectures is raised to accelerate the convergence process and improve the efficiency of the algorithm.Experimental studies show that the proposed algorithm has competitive performance comparable to many existing manual and automatic deep neural network generation approaches,which achieved a parameter-less and high-precision architecture with low-cost on each of the six benchmark datasets.
文摘In this paper,a novel method of ultra-lightweight convolution neural network(CNN)design based on neural architecture search(NAS)and knowledge distillation(KD)is proposed.It can realize the automatic construction of the space target inverse synthetic aperture radar(ISAR)image recognition model with ultra-lightweight and high accuracy.This method introduces the NAS method into the radar image recognition for the first time,which solves the time-consuming and labor-consuming problems in the artificial design of the space target ISAR image automatic recognition model(STIIARM).On this basis,the NAS model’s knowledge is transferred to the student model with lower computational complexity by the flow of the solution procedure(FSP)distillation method.Thus,the decline of recognition accuracy caused by the direct compression of model structural parameters can be effectively avoided,and the ultralightweight STIIARM can be obtained.In the method,the Inverted Linear Bottleneck(ILB)and Inverted Residual Block(IRB)are firstly taken as each block’s basic structure in CNN.And the expansion ratio,output filter size,number of IRBs,and convolution kernel size are set as the search parameters to construct a hierarchical decomposition search space.Then,the recognition accuracy and computational complexity are taken as the objective function and constraint conditions,respectively,and the global optimization model of the CNN architecture search is established.Next,the simulated annealing(SA)algorithm is used as the search strategy to search out the lightweight and high accuracy STIIARM directly.After that,based on the three principles of similar block structure,the same corresponding channel number,and the minimum computational complexity,the more lightweight student model is designed,and the FSP matrix pairing between the NAS model and student model is completed.Finally,by minimizing the loss between the FSP matrix pairs of the NAS model and student model,the student model’s weight adjustment is completed.Thus the ultra-lightweight and high accuracy STIIARM is obtained.The proposed method’s effectiveness is verified by the simulation experiments on the ISAR image dataset of five types of space targets.
基金This work was supported by a Ulucu PhD studentshipY.Jin is funded by an Alexander von Humboldt Professorship for Artificial Intelligence endowed by the German Federal Ministry of Education and Research.
文摘Neural architecture search(NAS)has become increasingly popular in the deep learning community recently,mainly because it can provide an opportunity to allow interested users without rich expertise to benefit from the success of deep neural networks(DNNs).However,NAS is still laborious and time-consuming because a large number of performance estimations are required during the search process of NAS,and training DNNs is computationally intensive.To solve this major limitation of NAS,improving the computational efficiency is essential in the design of NAS.However,a systematic overview of computationally efficient NAS(CE-NAS)methods still lacks.To fill this gap,we provide a comprehensive survey of the state-of-the-art on CE-NAS by categorizing the existing work into proxy-based and surrogate-assisted NAS methods,together with a thorough discussion of their design principles and a quantitative comparison of their performances and computational complexities.The remaining challenges and open research questions are also discussed,and promising research topics in this emerging field are suggested.
文摘AutoML(Automated Machine Learning)is an emerging field that aims to automate the process of building machine learning models.AutoML emerged to increase productivity and efficiency by automating as much as possible the inefficient work that occurs while repeating this process whenever machine learning is applied.In particular,research has been conducted for a long time on technologies that can effectively develop high-quality models by minimizing the intervention of model developers in the process from data preprocessing to algorithm selection and tuning.In this semantic review research,we summarize the data processing requirements for AutoML approaches and provide a detailed explanation.We place greater emphasis on neural architecture search(NAS)as it currently represents a highly popular sub-topic within the field of AutoML.NAS methods use machine learning algorithms to search through a large space of possible architectures and find the one that performs best on a given task.We provide a summary of the performance achieved by representative NAS algorithms on the CIFAR-10,CIFAR-100,ImageNet and wellknown benchmark datasets.Additionally,we delve into several noteworthy research directions in NAS methods including one/two-stage NAS,one-shot NAS and joint hyperparameter with architecture optimization.We discussed how the search space size and complexity in NAS can vary depending on the specific problem being addressed.To conclude,we examine several open problems(SOTA problems)within current AutoML methods that assure further investigation in future research.
基金supported by the Liaoning Province Applied Basic Research Program Project of China(Grant:2023JH2/101300065)the Liaoning Province Science and Technology Plan Joint Fund(2023-MSLH-221).
文摘Deep learning has emerged in many practical applications,such as image classification,fault diagnosis,and object detection.More recently,convolutional neural networks(CNNs),representative models of deep learning,have been used to solve fault detection.However,the current design of CNNs for fault detection of wind turbine blades is highly dependent on domain knowledge and requires a large amount of trial and error.For this reason,an evolutionary YOLOv8 network has been developed to automatically find the network architecture for wind turbine blade-based fault detection.YOLOv8 is a CNN-backed object detection model.Specifically,to reduce the parameter count,we first design an improved FasterNet module based on the Partial Convolution(PConv)operator.Then,to enhance convergence performance,we improve the loss function based on the efficient complete intersection over the union.Based on this,a flexible variable-length encoding is proposed,and the corresponding reproduction operators are designed.Related experimental results confirmthat the proposed approach can achieve better fault detection results and improve by 2.6%in mean precision at 50(mAP50)compared to the existing methods.Additionally,compared to training with the YOLOv8n model,the YOLOBFE model reduces the training parameters by 933,937 and decreases the GFLOPS(Giga Floating Point Operations Per Second)by 1.1.
