Abnormal network traffic, as a frequent security risk, requires a series of techniques to categorize and detect it. Existing network traffic anomaly detection still faces challenges: the inability to fully extract loc...Abnormal network traffic, as a frequent security risk, requires a series of techniques to categorize and detect it. Existing network traffic anomaly detection still faces challenges: the inability to fully extract local and global features, as well as the lack of effective mechanisms to capture complex interactions between features;Additionally, when increasing the receptive field to obtain deeper feature representations, the reliance on increasing network depth leads to a significant increase in computational resource consumption, affecting the efficiency and performance of detection. Based on these issues, firstly, this paper proposes a network traffic anomaly detection model based on parallel dilated convolution and residual learning (Res-PDC). To better explore the interactive relationships between features, the traffic samples are converted into two-dimensional matrix. A module combining parallel dilated convolutions and residual learning (res-pdc) was designed to extract local and global features of traffic at different scales. By utilizing res-pdc modules with different dilation rates, we can effectively capture spatial features at different scales and explore feature dependencies spanning wider regions without increasing computational resources. Secondly, to focus and integrate the information in different feature subspaces, further enhance and extract the interactions among the features, multi-head attention is added to Res-PDC, resulting in the final model: multi-head attention enhanced parallel dilated convolution and residual learning (MHA-Res-PDC) for network traffic anomaly detection. Finally, comparisons with other machine learning and deep learning algorithms are conducted on the NSL-KDD and CIC-IDS-2018 datasets. The experimental results demonstrate that the proposed method in this paper can effectively improve the detection performance.展开更多
The machine learning models of multiple linear regression(MLR),support vector regression(SVR),and extreme learning ma-chine(ELM)and the proposed ELM models of online sequential ELM(OS-ELM)and OS-ELM with forgetting me...The machine learning models of multiple linear regression(MLR),support vector regression(SVR),and extreme learning ma-chine(ELM)and the proposed ELM models of online sequential ELM(OS-ELM)and OS-ELM with forgetting mechanism(FOS-ELM)are applied in the prediction of the lime utilization ratio of dephosphorization in the basic oxygen furnace steelmaking process.The ELM model exhibites the best performance compared with the models of MLR and SVR.OS-ELM and FOS-ELM are applied for sequential learning and model updating.The optimal number of samples in validity term of the FOS-ELM model is determined to be 1500,with the smallest population mean absolute relative error(MARE)value of 0.058226 for the population.The variable importance analysis reveals lime weight,initial P content,and hot metal weight as the most important variables for the lime utilization ratio.The lime utilization ratio increases with the decrease in lime weight and the increases in the initial P content and hot metal weight.A prediction system based on FOS-ELM is applied in actual industrial production for one month.The hit ratios of the predicted lime utilization ratio in the error ranges of±1%,±3%,and±5%are 61.16%,90.63%,and 94.11%,respectively.The coefficient of determination,MARE,and root mean square error are 0.8670,0.06823,and 1.4265,respectively.The system exhibits desirable performance for applications in actual industrial pro-duction.展开更多
The optimization of process parameters in polyolefin production can bring significant economic benefits to the factory.However,due to small data sets,high costs associated with parameter verification cycles,and diffic...The optimization of process parameters in polyolefin production can bring significant economic benefits to the factory.However,due to small data sets,high costs associated with parameter verification cycles,and difficulty in establishing an optimization model,the optimization process is often restricted.To address this issue,we propose using a transfer learning Bayesian optimization strategy to improve the efficiency of parameter optimization while minimizing resource consumption.Specifically,we leverage Gaussian process(GP)regression models to establish an integrated model that incorporates both source and target grade production task data.We then measure the similarity weights of each model by comparing their predicted trends,and utilize these weights to accelerate the solution of optimal process parameters for producing target polyolefin grades.In order to enhance the accuracy of our approach,we acknowledge that measuring similarity in a global search space may not effectively capture local similarity characteristics.Therefore,we propose a novel method for transfer learning optimization that operates within a local space(LSTL-PBO).This method employs partial data acquired through random sampling from the target task data and utilizes Bayesian optimization techniques for model establishment.By focusing on a local search space,we aim to better discern and leverage the inherent similarities between source tasks and the target task.Additionally,we incorporate a parallel concept into our method to address multiple local search spaces simultaneously.By doing so,we can explore different regions of the parameter space in parallel,thereby increasing the chances of finding optimal process parameters.This localized approach allows us to improve the precision and effectiveness of our optimization process.The performance of our method is validated through experiments on benchmark problems,and we discuss the sensitivity of its hyperparameters.The results show that our proposed method can significantly improve the efficiency of process parameter optimization,reduce the dependence on source tasks,and enhance the method's robustness.This has great potential for optimizing processes in industrial environments.展开更多
Hyperparameter tuning is a key step in developing high-performing machine learning models, but searching large hyperparameter spaces requires extensive computation using standard sequential methods. This work analyzes...Hyperparameter tuning is a key step in developing high-performing machine learning models, but searching large hyperparameter spaces requires extensive computation using standard sequential methods. This work analyzes the performance gains from parallel versus sequential hyperparameter optimization. Using scikit-learn’s Randomized SearchCV, this project tuned a Random Forest classifier for fake news detection via randomized grid search. Setting n_jobs to -1 enabled full parallelization across CPU cores. Results show the parallel implementation achieved over 5× faster CPU times and 3× faster total run times compared to sequential tuning. However, test accuracy slightly dropped from 99.26% sequentially to 99.15% with parallelism, indicating a trade-off between evaluation efficiency and model performance. Still, the significant computational gains allow more extensive hyperparameter exploration within reasonable timeframes, outweighing the small accuracy decrease. Further analysis could better quantify this trade-off across different models, tuning techniques, tasks, and hardware.展开更多
Deep neural networks(DNNs)are effective in solving both forward and inverse problems for nonlinear partial differential equations(PDEs).However,conventional DNNs are not effective in handling problems such as delay di...Deep neural networks(DNNs)are effective in solving both forward and inverse problems for nonlinear partial differential equations(PDEs).However,conventional DNNs are not effective in handling problems such as delay differential equations(DDEs)and delay integrodifferential equations(DIDEs)with constant delays,primarily due to their low regularity at delayinduced breaking points.In this paper,a DNN method that combines multi-task learning(MTL)which is proposed to solve both the forward and inverse problems of DIDEs.The core idea of this approach is to divide the original equation into multiple tasks based on the delay,using auxiliary outputs to represent the integral terms,followed by the use of MTL to seamlessly incorporate the properties at the breaking points into the loss function.Furthermore,given the increased training dificulty associated with multiple tasks and outputs,we employ a sequential training scheme to reduce training complexity and provide reference solutions for subsequent tasks.This approach significantly enhances the approximation accuracy of solving DIDEs with DNNs,as demonstrated by comparisons with traditional DNN methods.We validate the effectiveness of this method through several numerical experiments,test various parameter sharing structures in MTL and compare the testing results of these structures.Finally,this method is implemented to solve the inverse problem of nonlinear DIDE and the results show that the unknown parameters of DIDE can be discovered with sparse or noisy data.展开更多
Propose the sequential circuits with the ternary D-ffs in series^1. Discuss the equivalence between the sequential circuits with the p-valued flip-flops in series and in parallel as a part of studying the multiple val...Propose the sequential circuits with the ternary D-ffs in series^1. Discuss the equivalence between the sequential circuits with the p-valued flip-flops in series and in parallel as a part of studying the multiple valued logic circuits.展开更多
The development of machine learning in complex system is hindered by two problems nowadays.The first problem is the inefficiency of exploration in state and action space,which leads to the data-hungry of some state-of...The development of machine learning in complex system is hindered by two problems nowadays.The first problem is the inefficiency of exploration in state and action space,which leads to the data-hungry of some state-of-art data-driven algorithm.The second problem is the lack of a general theory which can be used to analyze and implement a complex learning system.In this paper,we proposed a general methods that can address both two issues.We combine the concepts of descriptive learning,predictive learning,and prescriptive learning into a uniform framework,so as to build a parallel system allowing learning system improved by self-boosting.Formulating a new perspective of data,knowledge and action,we provide a new methodology called parallel learning to design machine learning system for real-world problems.展开更多
In this paper, a new machine learning framework is developed for complex system control, called parallel reinforcement learning. To overcome data deficiency of current data-driven algorithms, a parallel system is buil...In this paper, a new machine learning framework is developed for complex system control, called parallel reinforcement learning. To overcome data deficiency of current data-driven algorithms, a parallel system is built to improve complex learning system by self-guidance. Based on the Markov chain(MC) theory, we combine the transfer learning, predictive learning, deep learning and reinforcement learning to tackle the data and action processes and to express the knowledge. Parallel reinforcement learning framework is formulated and several case studies for real-world problems are finally introduced.展开更多
The virtual-to-real paradigm,i.e.,training models on virtual data and then applying them to solve real-world problems,has attracted more and more attention from various domains by successfully alleviating the data sho...The virtual-to-real paradigm,i.e.,training models on virtual data and then applying them to solve real-world problems,has attracted more and more attention from various domains by successfully alleviating the data shortage problem in machine learning.To summarize the advances in recent years,this survey comprehensively reviews the literature,from the viewport of parallel intelligence.First,an extended parallel learning framework is proposed to cover main domains including computer vision,natural language processing,robotics,and autonomous driving.Second,a multi-dimensional taxonomy is designed to organize the literature in a hierarchical structure.Third,the related virtual-toreal works are analyzed and compared according to the three principles of parallel learning known as description,prediction,and prescription,which cover the methods for constructing virtual worlds,generating labeled data,domain transferring,model training and testing,as well as optimizing the strategies to guide the task-oriented data generator for better learning performance.Key issues remained in virtual-to-real are discussed.Furthermore,the future research directions from the viewpoint of parallel learning are suggested.展开更多
As a complex and critical cyber-physical system(CPS),the hybrid electric powertrain is significant to mitigate air pollution and improve fuel economy.Energy management strategy(EMS)is playing a key role to improve the...As a complex and critical cyber-physical system(CPS),the hybrid electric powertrain is significant to mitigate air pollution and improve fuel economy.Energy management strategy(EMS)is playing a key role to improve the energy efficiency of this CPS.This paper presents a novel bidirectional long shortterm memory(LSTM)network based parallel reinforcement learning(PRL)approach to construct EMS for a hybrid tracked vehicle(HTV).This method contains two levels.The high-level establishes a parallel system first,which includes a real powertrain system and an artificial system.Then,the synthesized data from this parallel system is trained by a bidirectional LSTM network.The lower-level determines the optimal EMS using the trained action state function in the model-free reinforcement learning(RL)framework.PRL is a fully data-driven and learning-enabled approach that does not depend on any prediction and predefined rules.Finally,real vehicle testing is implemented and relevant experiment data is collected and calibrated.Experimental results validate that the proposed EMS can achieve considerable energy efficiency improvement by comparing with the conventional RL approach and deep RL.展开更多
Aiming at addressing the problem of manoeuvring decision-making in UAV air combat,this study establishes a one-to-one air combat model,defines missile attack areas,and uses the non-deterministic policy Soft-Actor-Crit...Aiming at addressing the problem of manoeuvring decision-making in UAV air combat,this study establishes a one-to-one air combat model,defines missile attack areas,and uses the non-deterministic policy Soft-Actor-Critic(SAC)algorithm in deep reinforcement learning to construct a decision model to realize the manoeuvring process.At the same time,the complexity of the proposed algorithm is calculated,and the stability of the closed-loop system of air combat decision-making controlled by neural network is analysed by the Lyapunov function.This study defines the UAV air combat process as a gaming process and proposes a Parallel Self-Play training SAC algorithm(PSP-SAC)to improve the generalisation performance of UAV control decisions.Simulation results have shown that the proposed algorithm can realize sample sharing and policy sharing in multiple combat environments and can significantly improve the generalisation ability of the model compared to independent training.展开更多
A recommender system(RS)relying on latent factor analysis usually adopts stochastic gradient descent(SGD)as its learning algorithm.However,owing to its serial mechanism,an SGD algorithm suffers from low efficiency and...A recommender system(RS)relying on latent factor analysis usually adopts stochastic gradient descent(SGD)as its learning algorithm.However,owing to its serial mechanism,an SGD algorithm suffers from low efficiency and scalability when handling large-scale industrial problems.Aiming at addressing this issue,this study proposes a momentum-incorporated parallel stochastic gradient descent(MPSGD)algorithm,whose main idea is two-fold:a)implementing parallelization via a novel datasplitting strategy,and b)accelerating convergence rate by integrating momentum effects into its training process.With it,an MPSGD-based latent factor(MLF)model is achieved,which is capable of performing efficient and high-quality recommendations.Experimental results on four high-dimensional and sparse matrices generated by industrial RS indicate that owing to an MPSGD algorithm,an MLF model outperforms the existing state-of-the-art ones in both computational efficiency and scalability.展开更多
A novel sequential neural network learning algorithm for function approximation is presented. The multi-step-ahead output predictor of the stochastic time series is introduced to the growing and pruning network for co...A novel sequential neural network learning algorithm for function approximation is presented. The multi-step-ahead output predictor of the stochastic time series is introduced to the growing and pruning network for constructing network structure. And the network parameters are adjusted by the proportional differential filter （PDF） rather than EKF when the network growing criteria are not met. Experimental results show that the proposed algorithm can obtain a more compact network along with a smaller error in mean square sense than other typical sequential learning algorithms.展开更多
Infectious keratitis is the most common condition of corneal diseases in which a pathogen grows in the cornea leading to inflammation and destruction of the corneal tissues.Infectious keratitis is a medical emergency ...Infectious keratitis is the most common condition of corneal diseases in which a pathogen grows in the cornea leading to inflammation and destruction of the corneal tissues.Infectious keratitis is a medical emergency for which a rapid and accurate diagnosis is needed to ensure prompt and precise treatment to halt the disease progression and to limit the extent of corneal damage;otherwise,it may develop a sight-threatening and even eye-globe-threatening condition.In this paper,we propose a sequentiallevel deep model to effectively discriminate infectious corneal disease via the classification of clinical images.In this approach,we devise an appropriate mechanism to preserve the spatial structures of clinical images and disentangle the informative features for clinical image classification of infectious keratitis.In a comparison,the performance of the proposed sequential-level deep model achieved 80%diagnostic accuracy,far better than the 49.27%±11.5%diagnostic accuracy achieved by 421 ophthalmologists over 120 test images.展开更多
To realize high-accuracy physical-cyber digital twin(DT)mapping in a manufacturing system,a huge amount of data need to be collected and analyzed in real-time.Traditional DTs systems are deployed in cloud or edge serv...To realize high-accuracy physical-cyber digital twin(DT)mapping in a manufacturing system,a huge amount of data need to be collected and analyzed in real-time.Traditional DTs systems are deployed in cloud or edge servers independently,whilst it is hard to apply in real production systems due to the high interaction or execution delay.This results in a low consistency in the temporal dimension of the physical-cyber model.In this work,we propose a novel efficient edge-cloud DT manufacturing system,which is inspired by resource scheduling technology.Specifically,an edge-cloud collaborative DTs system deployment architecture is first constructed.Then,deterministic and uncertainty optimization adaptive strategies are presented to choose a more powerful server for running DT-based applications.We model the adaptive optimization problems as dynamic programming problems and propose a novel collaborative clustering parallel Q-learning(CCPQL)algorithm and prediction-based CCPQL to solve the problems.The proposed approach reduces the total delay with a higher convergence rate.Numerical simulation results are provided to validate the approach,which would have great potential in dynamic and complex industrial internet environments.展开更多
A new parallel architecture for quantified boolean formula(QBF)solving was proposed,and the prediction model based on machine learning technology was proposed for how sharing knowledge affects the solving performance ...A new parallel architecture for quantified boolean formula(QBF)solving was proposed,and the prediction model based on machine learning technology was proposed for how sharing knowledge affects the solving performance in QBF parallel solving system,and the experimental evaluation scheme was also designed.It shows that the characterization factor of clause and cube influence the solving performance markedly in our experiment.At the same time,the heuristic machine learning algorithm was applied,support vector machine was chosen to predict the performance of QBF parallel solving system based on clause sharing and cube sharing.The relative error of accuracy for prediction can be controlled in a reasonable range of 20%30%.The results show the important and complex role that knowledge sharing plays in any modern parallel solver.It shows that the parallel solver with machine learning reduces the quantity of knowledge sharing about 30%and saving computational resource but does not reduce the performance of solving system.展开更多
Currently,energy conservation draws wide attention in industrial manufacturing systems.In recent years,many studies have aimed at saving energy consumption in the process of manufacturing and scheduling is regarded as...Currently,energy conservation draws wide attention in industrial manufacturing systems.In recent years,many studies have aimed at saving energy consumption in the process of manufacturing and scheduling is regarded as an effective approach.This paper puts forwards a multi-objective stochastic parallel machine scheduling problem with the consideration of deteriorating and learning effects.In it,the real processing time of jobs is calculated by using their processing speed and normal processing time.To describe this problem in a mathematical way,amultiobjective stochastic programming model aiming at realizing makespan and energy consumption minimization is formulated.Furthermore,we develop a multi-objective multi-verse optimization combined with a stochastic simulation method to deal with it.In this approach,the multi-verse optimization is adopted to find favorable solutions from the huge solution domain,while the stochastic simulation method is employed to assess them.By conducting comparison experiments on test problems,it can be verified that the developed approach has better performance in coping with the considered problem,compared to two classic multi-objective evolutionary algorithms.展开更多
In this paper, we consider scheduling problems with general truncated job-dependent learning effect on unrelated parallel-machine. The objective functions are to minimize total machine load, total completion (waiting)...In this paper, we consider scheduling problems with general truncated job-dependent learning effect on unrelated parallel-machine. The objective functions are to minimize total machine load, total completion (waiting) time, total absolute differences in completion (waiting) times respectively. If the number of machines is fixed, these problems can be solved in time respectively, where m is the number of machines and n is the number of jobs.展开更多
基金supported by the Xiamen Science and Technology Subsidy Project(No.2023CXY0318).
文摘Abnormal network traffic, as a frequent security risk, requires a series of techniques to categorize and detect it. Existing network traffic anomaly detection still faces challenges: the inability to fully extract local and global features, as well as the lack of effective mechanisms to capture complex interactions between features;Additionally, when increasing the receptive field to obtain deeper feature representations, the reliance on increasing network depth leads to a significant increase in computational resource consumption, affecting the efficiency and performance of detection. Based on these issues, firstly, this paper proposes a network traffic anomaly detection model based on parallel dilated convolution and residual learning (Res-PDC). To better explore the interactive relationships between features, the traffic samples are converted into two-dimensional matrix. A module combining parallel dilated convolutions and residual learning (res-pdc) was designed to extract local and global features of traffic at different scales. By utilizing res-pdc modules with different dilation rates, we can effectively capture spatial features at different scales and explore feature dependencies spanning wider regions without increasing computational resources. Secondly, to focus and integrate the information in different feature subspaces, further enhance and extract the interactions among the features, multi-head attention is added to Res-PDC, resulting in the final model: multi-head attention enhanced parallel dilated convolution and residual learning (MHA-Res-PDC) for network traffic anomaly detection. Finally, comparisons with other machine learning and deep learning algorithms are conducted on the NSL-KDD and CIC-IDS-2018 datasets. The experimental results demonstrate that the proposed method in this paper can effectively improve the detection performance.
基金supported by the National Natural Science Foundation of China (No.U1960202).
文摘The machine learning models of multiple linear regression(MLR),support vector regression(SVR),and extreme learning ma-chine(ELM)and the proposed ELM models of online sequential ELM(OS-ELM)and OS-ELM with forgetting mechanism(FOS-ELM)are applied in the prediction of the lime utilization ratio of dephosphorization in the basic oxygen furnace steelmaking process.The ELM model exhibites the best performance compared with the models of MLR and SVR.OS-ELM and FOS-ELM are applied for sequential learning and model updating.The optimal number of samples in validity term of the FOS-ELM model is determined to be 1500,with the smallest population mean absolute relative error(MARE)value of 0.058226 for the population.The variable importance analysis reveals lime weight,initial P content,and hot metal weight as the most important variables for the lime utilization ratio.The lime utilization ratio increases with the decrease in lime weight and the increases in the initial P content and hot metal weight.A prediction system based on FOS-ELM is applied in actual industrial production for one month.The hit ratios of the predicted lime utilization ratio in the error ranges of±1%,±3%,and±5%are 61.16%,90.63%,and 94.11%,respectively.The coefficient of determination,MARE,and root mean square error are 0.8670,0.06823,and 1.4265,respectively.The system exhibits desirable performance for applications in actual industrial pro-duction.
基金supported by National Natural Science Foundation of China(62394343)Major Program of Qingyuan Innovation Laboratory(00122002)+1 种基金Major Science and Technology Projects of Longmen Laboratory(231100220600)Shanghai Committee of Science and Technology(23ZR1416000)and Shanghai AI Lab.
文摘The optimization of process parameters in polyolefin production can bring significant economic benefits to the factory.However,due to small data sets,high costs associated with parameter verification cycles,and difficulty in establishing an optimization model,the optimization process is often restricted.To address this issue,we propose using a transfer learning Bayesian optimization strategy to improve the efficiency of parameter optimization while minimizing resource consumption.Specifically,we leverage Gaussian process(GP)regression models to establish an integrated model that incorporates both source and target grade production task data.We then measure the similarity weights of each model by comparing their predicted trends,and utilize these weights to accelerate the solution of optimal process parameters for producing target polyolefin grades.In order to enhance the accuracy of our approach,we acknowledge that measuring similarity in a global search space may not effectively capture local similarity characteristics.Therefore,we propose a novel method for transfer learning optimization that operates within a local space(LSTL-PBO).This method employs partial data acquired through random sampling from the target task data and utilizes Bayesian optimization techniques for model establishment.By focusing on a local search space,we aim to better discern and leverage the inherent similarities between source tasks and the target task.Additionally,we incorporate a parallel concept into our method to address multiple local search spaces simultaneously.By doing so,we can explore different regions of the parameter space in parallel,thereby increasing the chances of finding optimal process parameters.This localized approach allows us to improve the precision and effectiveness of our optimization process.The performance of our method is validated through experiments on benchmark problems,and we discuss the sensitivity of its hyperparameters.The results show that our proposed method can significantly improve the efficiency of process parameter optimization,reduce the dependence on source tasks,and enhance the method's robustness.This has great potential for optimizing processes in industrial environments.
文摘Hyperparameter tuning is a key step in developing high-performing machine learning models, but searching large hyperparameter spaces requires extensive computation using standard sequential methods. This work analyzes the performance gains from parallel versus sequential hyperparameter optimization. Using scikit-learn’s Randomized SearchCV, this project tuned a Random Forest classifier for fake news detection via randomized grid search. Setting n_jobs to -1 enabled full parallelization across CPU cores. Results show the parallel implementation achieved over 5× faster CPU times and 3× faster total run times compared to sequential tuning. However, test accuracy slightly dropped from 99.26% sequentially to 99.15% with parallelism, indicating a trade-off between evaluation efficiency and model performance. Still, the significant computational gains allow more extensive hyperparameter exploration within reasonable timeframes, outweighing the small accuracy decrease. Further analysis could better quantify this trade-off across different models, tuning techniques, tasks, and hardware.
文摘Deep neural networks(DNNs)are effective in solving both forward and inverse problems for nonlinear partial differential equations(PDEs).However,conventional DNNs are not effective in handling problems such as delay differential equations(DDEs)and delay integrodifferential equations(DIDEs)with constant delays,primarily due to their low regularity at delayinduced breaking points.In this paper,a DNN method that combines multi-task learning(MTL)which is proposed to solve both the forward and inverse problems of DIDEs.The core idea of this approach is to divide the original equation into multiple tasks based on the delay,using auxiliary outputs to represent the integral terms,followed by the use of MTL to seamlessly incorporate the properties at the breaking points into the loss function.Furthermore,given the increased training dificulty associated with multiple tasks and outputs,we employ a sequential training scheme to reduce training complexity and provide reference solutions for subsequent tasks.This approach significantly enhances the approximation accuracy of solving DIDEs with DNNs,as demonstrated by comparisons with traditional DNN methods.We validate the effectiveness of this method through several numerical experiments,test various parameter sharing structures in MTL and compare the testing results of these structures.Finally,this method is implemented to solve the inverse problem of nonlinear DIDE and the results show that the unknown parameters of DIDE can be discovered with sparse or noisy data.
文摘Propose the sequential circuits with the ternary D-ffs in series^1. Discuss the equivalence between the sequential circuits with the p-valued flip-flops in series and in parallel as a part of studying the multiple valued logic circuits.
基金supported in part by the National Natural Science Foundation of China(91520301)
文摘The development of machine learning in complex system is hindered by two problems nowadays.The first problem is the inefficiency of exploration in state and action space,which leads to the data-hungry of some state-of-art data-driven algorithm.The second problem is the lack of a general theory which can be used to analyze and implement a complex learning system.In this paper,we proposed a general methods that can address both two issues.We combine the concepts of descriptive learning,predictive learning,and prescriptive learning into a uniform framework,so as to build a parallel system allowing learning system improved by self-boosting.Formulating a new perspective of data,knowledge and action,we provide a new methodology called parallel learning to design machine learning system for real-world problems.
基金supported in part by the National Natural Science Foundation of China(61503380)the Natural Science Foundation of Guangdong Province,China(2015A030310187)
文摘In this paper, a new machine learning framework is developed for complex system control, called parallel reinforcement learning. To overcome data deficiency of current data-driven algorithms, a parallel system is built to improve complex learning system by self-guidance. Based on the Markov chain(MC) theory, we combine the transfer learning, predictive learning, deep learning and reinforcement learning to tackle the data and action processes and to express the knowledge. Parallel reinforcement learning framework is formulated and several case studies for real-world problems are finally introduced.
基金partially supported by the National Key Research and Development Program of China(2020YFB2104001)the National Natural Science Foundation of China(62271485,61903363,U1811463)Open Project of the State Key Laboratory for Management and Control of Complex Systems(20220117)。
文摘The virtual-to-real paradigm,i.e.,training models on virtual data and then applying them to solve real-world problems,has attracted more and more attention from various domains by successfully alleviating the data shortage problem in machine learning.To summarize the advances in recent years,this survey comprehensively reviews the literature,from the viewport of parallel intelligence.First,an extended parallel learning framework is proposed to cover main domains including computer vision,natural language processing,robotics,and autonomous driving.Second,a multi-dimensional taxonomy is designed to organize the literature in a hierarchical structure.Third,the related virtual-toreal works are analyzed and compared according to the three principles of parallel learning known as description,prediction,and prescription,which cover the methods for constructing virtual worlds,generating labeled data,domain transferring,model training and testing,as well as optimizing the strategies to guide the task-oriented data generator for better learning performance.Key issues remained in virtual-to-real are discussed.Furthermore,the future research directions from the viewpoint of parallel learning are suggested.
基金supported in part by the National Natural Science Foundation of China(61533019,91720000)Beijing Municipal Science and Technology Commission(Z181100008918007)the Intel Collaborative Research Institute for Intelligent and Automated Connected Vehicles(pICRI-IACVq)
文摘As a complex and critical cyber-physical system(CPS),the hybrid electric powertrain is significant to mitigate air pollution and improve fuel economy.Energy management strategy(EMS)is playing a key role to improve the energy efficiency of this CPS.This paper presents a novel bidirectional long shortterm memory(LSTM)network based parallel reinforcement learning(PRL)approach to construct EMS for a hybrid tracked vehicle(HTV).This method contains two levels.The high-level establishes a parallel system first,which includes a real powertrain system and an artificial system.Then,the synthesized data from this parallel system is trained by a bidirectional LSTM network.The lower-level determines the optimal EMS using the trained action state function in the model-free reinforcement learning(RL)framework.PRL is a fully data-driven and learning-enabled approach that does not depend on any prediction and predefined rules.Finally,real vehicle testing is implemented and relevant experiment data is collected and calibrated.Experimental results validate that the proposed EMS can achieve considerable energy efficiency improvement by comparing with the conventional RL approach and deep RL.
基金National Natural Science Foundation of China,Grant/Award Number:62003267Fundamental Research Funds for the Central Universities,Grant/Award Number:G2022KY0602+1 种基金Technology on Electromagnetic Space Operations and Applications Laboratory,Grant/Award Number:2022ZX0090Key Core Technology Research Plan of Xi'an,Grant/Award Number:21RGZN0016。
文摘Aiming at addressing the problem of manoeuvring decision-making in UAV air combat,this study establishes a one-to-one air combat model,defines missile attack areas,and uses the non-deterministic policy Soft-Actor-Critic(SAC)algorithm in deep reinforcement learning to construct a decision model to realize the manoeuvring process.At the same time,the complexity of the proposed algorithm is calculated,and the stability of the closed-loop system of air combat decision-making controlled by neural network is analysed by the Lyapunov function.This study defines the UAV air combat process as a gaming process and proposes a Parallel Self-Play training SAC algorithm(PSP-SAC)to improve the generalisation performance of UAV control decisions.Simulation results have shown that the proposed algorithm can realize sample sharing and policy sharing in multiple combat environments and can significantly improve the generalisation ability of the model compared to independent training.
基金supported in part by the National Natural Science Foundation of China(61772493)the Deanship of Scientific Research(DSR)at King Abdulaziz University(RG-48-135-40)+1 种基金Guangdong Province Universities and College Pearl River Scholar Funded Scheme(2019)the Natural Science Foundation of Chongqing(cstc2019jcyjjqX0013)。
文摘A recommender system(RS)relying on latent factor analysis usually adopts stochastic gradient descent(SGD)as its learning algorithm.However,owing to its serial mechanism,an SGD algorithm suffers from low efficiency and scalability when handling large-scale industrial problems.Aiming at addressing this issue,this study proposes a momentum-incorporated parallel stochastic gradient descent(MPSGD)algorithm,whose main idea is two-fold:a)implementing parallelization via a novel datasplitting strategy,and b)accelerating convergence rate by integrating momentum effects into its training process.With it,an MPSGD-based latent factor(MLF)model is achieved,which is capable of performing efficient and high-quality recommendations.Experimental results on four high-dimensional and sparse matrices generated by industrial RS indicate that owing to an MPSGD algorithm,an MLF model outperforms the existing state-of-the-art ones in both computational efficiency and scalability.
基金Sponsored by the Ministerial Level Foundation(230032)
文摘A novel sequential neural network learning algorithm for function approximation is presented. The multi-step-ahead output predictor of the stochastic time series is introduced to the growing and pruning network for constructing network structure. And the network parameters are adjusted by the proportional differential filter （PDF） rather than EKF when the network growing criteria are not met. Experimental results show that the proposed algorithm can obtain a more compact network along with a smaller error in mean square sense than other typical sequential learning algorithms.
基金supported by the Health Commission of Zhejiang Province(WKJ-ZJ-1905 and 2018ZD007)the Key Research and Development Projects of Zhejiang Province(2018C03082)the National Natural Science Foundation of China(61625107)。
文摘Infectious keratitis is the most common condition of corneal diseases in which a pathogen grows in the cornea leading to inflammation and destruction of the corneal tissues.Infectious keratitis is a medical emergency for which a rapid and accurate diagnosis is needed to ensure prompt and precise treatment to halt the disease progression and to limit the extent of corneal damage;otherwise,it may develop a sight-threatening and even eye-globe-threatening condition.In this paper,we propose a sequentiallevel deep model to effectively discriminate infectious corneal disease via the classification of clinical images.In this approach,we devise an appropriate mechanism to preserve the spatial structures of clinical images and disentangle the informative features for clinical image classification of infectious keratitis.In a comparison,the performance of the proposed sequential-level deep model achieved 80%diagnostic accuracy,far better than the 49.27%±11.5%diagnostic accuracy achieved by 421 ophthalmologists over 120 test images.
基金supported by 2019 Industrial Internet Innovation Development Project of Ministry of Industry and Information Technology of P.R. China “Comprehensive Security Defense Platform Project for Industrial/Enterprise Networks”Research on Key Technologies of wireless edge intelligent collaboration for industrial internet scenarios (L202017)+1 种基金Natural Science Foundation of China, No.61971050BUPT Excellent Ph.D. Students Foundation (CX2020214)。
文摘To realize high-accuracy physical-cyber digital twin(DT)mapping in a manufacturing system,a huge amount of data need to be collected and analyzed in real-time.Traditional DTs systems are deployed in cloud or edge servers independently,whilst it is hard to apply in real production systems due to the high interaction or execution delay.This results in a low consistency in the temporal dimension of the physical-cyber model.In this work,we propose a novel efficient edge-cloud DT manufacturing system,which is inspired by resource scheduling technology.Specifically,an edge-cloud collaborative DTs system deployment architecture is first constructed.Then,deterministic and uncertainty optimization adaptive strategies are presented to choose a more powerful server for running DT-based applications.We model the adaptive optimization problems as dynamic programming problems and propose a novel collaborative clustering parallel Q-learning(CCPQL)algorithm and prediction-based CCPQL to solve the problems.The proposed approach reduces the total delay with a higher convergence rate.Numerical simulation results are provided to validate the approach,which would have great potential in dynamic and complex industrial internet environments.
基金Supported by National Natural Science Foundation of China（60474035）,National Research Foundation for the Doctoral Program of Higher Education of China（20050359004）,Natural Science Foundation of Anhui Province（070412035）
基金Manuscript received March 5, 2010 accepted March 2, 2011 Supported by National Natural Science Foundation of China （61004103）, National Research Foundation for the Doctoral Program of Higher Education of China （20100111110005）, China Postdoctoral Science Foundation （20090460742）, and Natural Science Foundation of Anhui Province of China （090412058, 11040606Q44）
基金Project(61171141)supported by the National Natural Science Foundation of China
文摘A new parallel architecture for quantified boolean formula(QBF)solving was proposed,and the prediction model based on machine learning technology was proposed for how sharing knowledge affects the solving performance in QBF parallel solving system,and the experimental evaluation scheme was also designed.It shows that the characterization factor of clause and cube influence the solving performance markedly in our experiment.At the same time,the heuristic machine learning algorithm was applied,support vector machine was chosen to predict the performance of QBF parallel solving system based on clause sharing and cube sharing.The relative error of accuracy for prediction can be controlled in a reasonable range of 20%30%.The results show the important and complex role that knowledge sharing plays in any modern parallel solver.It shows that the parallel solver with machine learning reduces the quantity of knowledge sharing about 30%and saving computational resource but does not reduce the performance of solving system.
文摘Currently,energy conservation draws wide attention in industrial manufacturing systems.In recent years,many studies have aimed at saving energy consumption in the process of manufacturing and scheduling is regarded as an effective approach.This paper puts forwards a multi-objective stochastic parallel machine scheduling problem with the consideration of deteriorating and learning effects.In it,the real processing time of jobs is calculated by using their processing speed and normal processing time.To describe this problem in a mathematical way,amultiobjective stochastic programming model aiming at realizing makespan and energy consumption minimization is formulated.Furthermore,we develop a multi-objective multi-verse optimization combined with a stochastic simulation method to deal with it.In this approach,the multi-verse optimization is adopted to find favorable solutions from the huge solution domain,while the stochastic simulation method is employed to assess them.By conducting comparison experiments on test problems,it can be verified that the developed approach has better performance in coping with the considered problem,compared to two classic multi-objective evolutionary algorithms.
文摘In this paper, we consider scheduling problems with general truncated job-dependent learning effect on unrelated parallel-machine. The objective functions are to minimize total machine load, total completion (waiting) time, total absolute differences in completion (waiting) times respectively. If the number of machines is fixed, these problems can be solved in time respectively, where m is the number of machines and n is the number of jobs.
