Traditional Chinese Medicine（TCM）, a crucial component of the current medical system, has been extensively used in clinical practice due to its valuable therapeutic efficacy, and its potentials as an important sourc...Traditional Chinese Medicine（TCM）, a crucial component of the current medical system, has been extensively used in clinical practice due to its valuable therapeutic efficacy, and its potentials as an important source of new pharmacophores. TCM is characterized by holistic theory, which emphasizes maintaining the balance of the patient＇s whole body using herbal formulae（fangji in Chinese） composed of mixtures of herbs with multiple bioactive ingredients. Because of the complex nature of these formulae, it is necessary to develop systematic methods to identify their bioactive ingredients and to clarify their mechanisms of action. With the rapid progress in bioinformatics, systems biology, and polypharmacology, ＂network pharmacology＂, which shifts the ＂one target, one drug＂ paradigm to the ＂network target, multi-component＂ strategy, has attracted the attention because it can not only reveal the underlying complex interactions between a herbal formula and cellular proteins but detect the influence of their interactions on the function and behavior of the system. Growing evidence shows that the network pharmacology strategy can be a powerful approach to modern research on TCM. The present paper focuses on the basis of network pharmacology and the recent progress in its methodology, illustrates its utility in screening bioactive ingredients and elucidating the mechanisms of action of TCM herbal formulae, analyzes its limitations and problems, and discusses its development direction and application prospects.展开更多
Traditional Chinese medicine(TCM)is a precious treasure of the Chinese nation and has unique advantages in the prevention and treatment of diseases.The holistic view of TCM coincides with the new generation of medical...Traditional Chinese medicine(TCM)is a precious treasure of the Chinese nation and has unique advantages in the prevention and treatment of diseases.The holistic view of TCM coincides with the new generation of medical research paradigm characterized by network and system.TCM gave birth to a new method featuring holistic and systematic"network target",a core theory and method of network pharmacology.TCM is also an important research object of network pharmacology.TCM network pharmacology,which aims to understand the network-based biological basis of complex diseases,TCM syndromes and herb treatments,plays a critical role in the origin and development process of network pharmacology.This review introduces new progresses of TCM network pharmacology in recent years,including predicting herb targets,understanding biological foundation of diseases and syndromes,network regulation mechanisms of herbal formulae,and identifying disease and syndrome biomarkers based on biological network.These studies show a trend of combining computational,experimental and clinical approaches,which is a promising direction of TCM network pharmacology research in the future.Considering that TCM network pharmacology is still a young research field,it is necessary to further standardize the research process and evaluation indicators to promote its healthy development.展开更多
Machine learning(ML) has been widely applied to the upper layers of wireless communication systems for various purposes, such as deployment of cognitive radio and communication network. However, its application to the...Machine learning(ML) has been widely applied to the upper layers of wireless communication systems for various purposes, such as deployment of cognitive radio and communication network. However, its application to the physical layer is hampered by sophisticated channel environments and limited learning ability of conventional ML algorithms. Deep learning(DL) has been recently applied for many fields, such as computer vision and natural language processing, given its expressive capacity and convenient optimization capability. The potential application of DL to the physical layer has also been increasingly recognized because of the new features for future communications, such as complex scenarios with unknown channel models, high speed and accurate processing requirements; these features challenge conventional communication theories. This paper presents a comprehensive overview of the emerging studies on DL-based physical layer processing, including leveraging DL to redesign a module of the conventional communication system(for modulation recognition, channel decoding, and detection) and replace the communication system with a radically new architecture based on an autoencoder. These DL-based methods show promising performance improvements but have certain limitations, such as lack of solid analytical tools and use of architectures that are specifically designed for communication and implementation research, thereby motivating future research in this field.展开更多
Foreword Note:Some of the document contents may involve certain patents,the identification of which is not the responsibility of the institution that releases the document.Main drafting organizations:Tsinghua Universi...Foreword Note:Some of the document contents may involve certain patents,the identification of which is not the responsibility of the institution that releases the document.Main drafting organizations:Tsinghua University,Specialty Committee of Network Pharmacology of World Federation of Chinese Medicine Societies(WFCMS).展开更多
The emerging development of connected and automated vehicles imposes a significant challenge on current vehicle control and transportation systems. This paper proposes a novel unified approach, Parallel Driving, a clo...The emerging development of connected and automated vehicles imposes a significant challenge on current vehicle control and transportation systems. This paper proposes a novel unified approach, Parallel Driving, a cloud-based cyberphysical-social systems(CPSS) framework aiming at synergizing connected automated driving. This study first introduces the CPSS and ACP-based intelligent machine systems. Then the parallel driving is proposed in the cyber-physical-social space,considering interactions among vehicles, human drivers, and information. Within the framework, parallel testing, parallel learning and parallel reinforcement learning are developed and concisely reviewed. Development on intelligent horizon(iHorizon)and its applications are also presented towards parallel horizon.The proposed parallel driving offers an ample solution for achieving a smooth, safe and efficient cooperation among connected automated vehicles with different levels of automation in future road transportation systems.展开更多
In this paper we numerically investigate the chaotic behaviours of the fractional-order Ikeda delay system. The results show that chaos exists in the fractional-order Ikeda delay system with order less than 1. The low...In this paper we numerically investigate the chaotic behaviours of the fractional-order Ikeda delay system. The results show that chaos exists in the fractional-order Ikeda delay system with order less than 1. The lowest order for chaos to be a, ble to appear in this system is found to be 0.1. Master-slave synchronization of chaotic fractional-order Ikeda delay systems with linear coupling is also studied.展开更多
Traditional Chinese medicine（TCM） holds a holistic theory, and specializes in balancing disordered human body using numerous natural products, particularly Chinese herbal formulae. TCM has certain treatment advantag...Traditional Chinese medicine（TCM） holds a holistic theory, and specializes in balancing disordered human body using numerous natural products, particularly Chinese herbal formulae. TCM has certain treatment advantages for patients suffering from various complex diseases. However, due to the complex nature of TCM, it remains difficult to unveil such holistic medicine by the current reductionism research strategies, which treat both herbal ingredients and targets in isolation. Recently, an emerging network pharmacology approach has been introduced to tackle this bottleneck problem. A TCM-derived novel therapeutic concept, ＂network target＂, which is different from the Western medicine＇s ＂onetarget＂ concept, has been proposed from China. The network target strategy is able to illustrate the complex interactions among the biological systems, drugs, and complex diseases from a network perspective, and thus provides an innovative approach to access ancient remedies in a precision manner and at a systematic level, which also highlights TCM＇s potential in current medical systems.展开更多
Smartphones and mobile tablets are rapidly becoming indispensable in daily life. Android has been the most popular mobile operating system since 2012. However, owing to the open nature of Android, countless malwares a...Smartphones and mobile tablets are rapidly becoming indispensable in daily life. Android has been the most popular mobile operating system since 2012. However, owing to the open nature of Android, countless malwares are hidden in a large number of benign apps in Android markets that seriously threaten Android security. Deep learning is a new area of machine learning research that has gained increasing attention in artificial intelligence. In this study, we propose to associate the features from the static analysis with features from dynamic analysis of Android apps and characterize malware using deep learning techniques. We implement an online deep-learning-based Android malware detection engine（Droid Detector） that can automatically detect whether an app is a malware or not. With thousands of Android apps, we thoroughly test Droid Detector and perform an indepth analysis on the features that deep learning essentially exploits to characterize malware. The results show that deep learning is suitable for characterizing Android malware and especially effective with the availability of more training data. Droid Detector can achieve 96.76% detection accuracy, which outperforms traditional machine learning techniques. An evaluation of ten popular anti-virus softwares demonstrates the urgency of advancing our capabilities in Android malware detection.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
Flexible job shop scheduling problems(FJSP)have received much attention from academia and industry for many years.Due to their exponential complexity,swarm intelligence(SI)and evolutionary algorithms(EA)are developed,...Flexible job shop scheduling problems(FJSP)have received much attention from academia and industry for many years.Due to their exponential complexity,swarm intelligence(SI)and evolutionary algorithms(EA)are developed,employed and improved for solving them.More than 60%of the publications are related to SI and EA.This paper intents to give a comprehensive literature review of SI and EA for solving FJSP.First,the mathematical model of FJSP is presented and the constraints in applications are summarized.Then,the encoding and decoding strategies for connecting the problem and algorithms are reviewed.The strategies for initializing algorithms?population and local search operators for improving convergence performance are summarized.Next,one classical hybrid genetic algorithm(GA)and one newest imperialist competitive algorithm(ICA)with variables neighborhood search(VNS)for solving FJSP are presented.Finally,we summarize,discus and analyze the status of SI and EA for solving FJSP and give insight into future research directions.展开更多
Lithium-ion batteries have become the third-generation space batteries and are widely utilized in a series of spacecraft. Remaining Useful Life （RUL） estimation is essential to a spacecraft as the battery is a criti...Lithium-ion batteries have become the third-generation space batteries and are widely utilized in a series of spacecraft. Remaining Useful Life （RUL） estimation is essential to a spacecraft as the battery is a critical part and determines the lifetime and reliability. The Relevance Vector Machine （RVM） is a data-driven algorithm used to estimate a battery＇s RUL due to its sparse feature and uncertainty management capability. Especially, some of the regressive cases indicate that the RVM can obtain a better short-term prediction performance rather than long-term prediction. As a nonlinear kernel learning algorithm, the coefficient matrix and relevance vectors are fixed once the RVM training is conducted. Moreover, the RVM can be simply influenced by the noise with the training data. Thus, this work proposes an iterative updated approach to improve the long-term prediction performance for a battery＇s RUL prediction. Firstly, when a new estimator is output by the RVM, the Kalman filter is applied to optimize this estimator with a physical degradation model. Then, this optimized estimator is added into the training set as an on-line sample, the RVM model is re-trained, and the coefficient matrix and relevance vectors can be dynamically adjusted to make next iterative prediction. Experimental results with a commercial battery test data set and a satellite battery data set both indicate that the proposed method can achieve a better performance for RUL estimation.展开更多
The application of intelligence to manufacturing has emerged as a compelling topic for researchers and industries around the world.However,different terminologies,namely smart manufacturing(SM)and intelligent manufact...The application of intelligence to manufacturing has emerged as a compelling topic for researchers and industries around the world.However,different terminologies,namely smart manufacturing(SM)and intelligent manufacturing(IM),have been applied to what may be broadly characterized as a similar paradigm by some researchers and practitioners.While SM and IM are similar,they are not identical.From an evolutionary perspective,there has been little consideration on whether the definition,thought,connotation,and technical development of the concepts of SM or IM are consistent in the literature.To address this gap,the work performs a qualitative and quantitative investigation of research literature to systematically compare inherent differences of SM and IM and clarify the relationship between SM and IM.A bibliometric analysis of publication sources,annual publication numbers,keyword frequency,and top regions of research and development establishes the scope and trends of the currently presented research.Critical topics discussed include origin,definitions,evolutionary path,and key technologies of SM and IM.The implementation architecture,standards,and national focus are also discussed.In this work,a basis to understand SM and IM is provided,which is increasingly important because the trend to merge both terminologies rises in Industry 4.0 as intelligence is being rapidly applied to modern manufacturing and human–cyber–physical systems.展开更多
Cross-media analysis and reasoning is an active research area in computer science, and a promising direction for artificial intelligence. However, to the best of our knowledge, no existing work has summarized the stat...Cross-media analysis and reasoning is an active research area in computer science, and a promising direction for artificial intelligence. However, to the best of our knowledge, no existing work has summarized the state-of-the-art methods for cross-media analysis and reasoning or presented advances, challenges, and future directions for the field. To address these issues, we provide an overview as follows： （1） theory and model for cross-media uniform representation; （2） cross-media correlation understanding and deep mining; （3） cross-media knowledge graph construction and learning methodologies; （4） cross-media knowledge evolution and reasoning; （5） cross-media description and generation; （6） cross-media intelligent engines; and （7） cross-media intelligent applications. By presenting approaches, advances, and future directions in cross-media analysis and reasoning, our goal is not only to draw more attention to the state-of-the-art advances in the field, but also to provide technical insights by discussing the challenges and research directions in these areas.展开更多
Capacitated vehicle routing problem (CVRP) is an NP-hard problem. For large-scale problems, it is quite difficult to achieve an optimal solution with traditional optimization methods due to the high computational comp...Capacitated vehicle routing problem (CVRP) is an NP-hard problem. For large-scale problems, it is quite difficult to achieve an optimal solution with traditional optimization methods due to the high computational complexity. A new hybrid ap- proximation algorithm is developed in this work to solve the problem. In the hybrid algorithm, discrete particle swarm optimiza- tion (DPSO) combines global search and local search to search for the optimal results and simulated annealing (SA) uses certain probability to avoid being trapped in a local optimum. The computational study showed that the proposed algorithm is a feasible and effective approach for capacitated vehicle routing problem, especially for large scale problems.展开更多
The development of gastritis is associated with an increased risk of gastric cancer. Current invasive gastritis diagnostic methods are not suitable for monitoring progressIn this work based on 78 gastritis patients an...The development of gastritis is associated with an increased risk of gastric cancer. Current invasive gastritis diagnostic methods are not suitable for monitoring progressIn this work based on 78 gastritis patients and 50 healthy individuals, we observed that the variation of tongue-coating microbiota was associated with the occurrenee and development of gastritis. Twenty-one microbial species were identified for differentiating tongue-coating microbiomes of gastritis and healthy individuals. Pathways such as microbial metabolism in diverse environments, biosynthesis of antibiotics and bacterial chemotaxis were up-regulated in gastritis patients. The abundance of Campylobacter concisus was found associated with the gastric precancerous cascade. Furthermore, Campylobacter concisus could be detected in tongue coating and gastric fluid in a validation cohort containing 38 gastritis patients. These observations provided biological evidence of tongue diagnosis in traditional Chinese medicine, and indicated that tongue-coating microbiome could be a potential non-invasive biomarker, which might be suitable for long-term monitoring of gastritis.展开更多
Background and purpose We investigated the baseline demographics of patients with severe unilateral atherosclerotic stenosis of the middle cerebral artery(MCA)using multimodal MRI and evaluated the haemodynamic impair...Background and purpose We investigated the baseline demographics of patients with severe unilateral atherosclerotic stenosis of the middle cerebral artery(MCA)using multimodal MRI and evaluated the haemodynamic impairments and plaque characteristics of patients who had a recurrent stroke.Materials and methods We retrospectively recruited consecutive patients with severe unilateral atherosclerotic MCA stenosis who underwent arterial spin labelling(ASL)with postlabelling delay(PLD)of 1.5 and 2.5 s,and vessel wall MRI.For each PLD,cerebral blood flow(CBF)maps were generated.Hypoperfusion volume ratio(HVR)from 2 PLD CBF was calculated.An HVR value≥50%was considered as severe HVR.Plaque areas,plaque burden,plaque length and remodelling index were measured.Plaque enhancement at maximal lumen narrowing site were graded.Baseline clinical and imaging characteristics were compared between patients with(event+)and without(event?)1 year ischaemic events.Results Forty-three patients(47.23±12.15 years;28 men)were enrolled in this study.Seven patients had an HVR≥50%.During the 1-year follow-up,7 patients had experienced a recurrent stroke.HVR were significantly higher in the event+than event?(53.17%±29.82%vs 16.9%±15.57%,p=0.0002),whereas no significant difference was detected in plaque areas,plaque burden,remodelling index,plaque length and plaque enhancement grade.The multivariable analysis revealed that a severe HVR was significantly associated with a recurrent stroke(Odds ratio=12.93,95%confidence interval 1.57 to 106.24,p=0.017)after adjusted by hypertension and smoking.Conclusion HVR obtained from two PLD ASL may be a useful imaging predictor of recurrent stroke.展开更多
CF3I has been widely considered as a potential alternative for SF6,because it has low global warming potential(GWP)but an insulation capability that is 1.2 times greater than that of SF6.In this paper,the electron swa...CF3I has been widely considered as a potential alternative for SF6,because it has low global warming potential(GWP)but an insulation capability that is 1.2 times greater than that of SF6.In this paper,the electron swarm parameters of CF3I and its gas mixture with N2,including the effective ionization coefficient and electron drift velocity,are examined theoretically through the Boltzmann equation method in the condition of steady-state Townsend(SST)experiments.Based on the derived data of the limiting field strength of CF3I-N2gas mixture,taking into consideration of environmental aspects such as GWP,ozone depletion potential(ODP),liquefaction temperature and toxicity,we studied the possibility of applying the gas mixture as the insulation medium in gas-insulated switchgears(GIS)or cubic type gas-insulated switchgears(C-GIS).It is found that CF3I-N2gas mixtures contained 30%～70%CF3I perform comprehensively better than pure SF6and compressed N2,and especially in medium and low voltage environments,the boiling point of CF3I-N2gas mixture meets the domestic and global requirements in mid-low latitude regions.Therefore we conclude that the gas mixture is acceptable for replacing SF6as the insulation medium in C-GISs.展开更多
基金National Natural Science Foundation of China(81225025)Beijing Nova Program(Z1511000003150126)
文摘Traditional Chinese Medicine（TCM）, a crucial component of the current medical system, has been extensively used in clinical practice due to its valuable therapeutic efficacy, and its potentials as an important source of new pharmacophores. TCM is characterized by holistic theory, which emphasizes maintaining the balance of the patient＇s whole body using herbal formulae（fangji in Chinese） composed of mixtures of herbs with multiple bioactive ingredients. Because of the complex nature of these formulae, it is necessary to develop systematic methods to identify their bioactive ingredients and to clarify their mechanisms of action. With the rapid progress in bioinformatics, systems biology, and polypharmacology, ＂network pharmacology＂, which shifts the ＂one target, one drug＂ paradigm to the ＂network target, multi-component＂ strategy, has attracted the attention because it can not only reveal the underlying complex interactions between a herbal formula and cellular proteins but detect the influence of their interactions on the function and behavior of the system. Growing evidence shows that the network pharmacology strategy can be a powerful approach to modern research on TCM. The present paper focuses on the basis of network pharmacology and the recent progress in its methodology, illustrates its utility in screening bioactive ingredients and elucidating the mechanisms of action of TCM herbal formulae, analyzes its limitations and problems, and discusses its development direction and application prospects.
基金supported by the National Natural Science Foundation of China(Nos.6201101081,81630103 and 81225025)Tsinghua University Spring Breeze Fund(No.2020-Z99CFY040)Beijing National Research Center for Information Science and Technology(Nos.BNR2019TD01020 and BNR2019-RC01012)。
文摘Traditional Chinese medicine(TCM)is a precious treasure of the Chinese nation and has unique advantages in the prevention and treatment of diseases.The holistic view of TCM coincides with the new generation of medical research paradigm characterized by network and system.TCM gave birth to a new method featuring holistic and systematic"network target",a core theory and method of network pharmacology.TCM is also an important research object of network pharmacology.TCM network pharmacology,which aims to understand the network-based biological basis of complex diseases,TCM syndromes and herb treatments,plays a critical role in the origin and development process of network pharmacology.This review introduces new progresses of TCM network pharmacology in recent years,including predicting herb targets,understanding biological foundation of diseases and syndromes,network regulation mechanisms of herbal formulae,and identifying disease and syndrome biomarkers based on biological network.These studies show a trend of combining computational,experimental and clinical approaches,which is a promising direction of TCM network pharmacology research in the future.Considering that TCM network pharmacology is still a young research field,it is necessary to further standardize the research process and evaluation indicators to promote its healthy development.
文摘Machine learning(ML) has been widely applied to the upper layers of wireless communication systems for various purposes, such as deployment of cognitive radio and communication network. However, its application to the physical layer is hampered by sophisticated channel environments and limited learning ability of conventional ML algorithms. Deep learning(DL) has been recently applied for many fields, such as computer vision and natural language processing, given its expressive capacity and convenient optimization capability. The potential application of DL to the physical layer has also been increasingly recognized because of the new features for future communications, such as complex scenarios with unknown channel models, high speed and accurate processing requirements; these features challenge conventional communication theories. This paper presents a comprehensive overview of the emerging studies on DL-based physical layer processing, including leveraging DL to redesign a module of the conventional communication system(for modulation recognition, channel decoding, and detection) and replace the communication system with a radically new architecture based on an autoencoder. These DL-based methods show promising performance improvements but have certain limitations, such as lack of solid analytical tools and use of architectures that are specifically designed for communication and implementation research, thereby motivating future research in this field.
文摘Foreword Note:Some of the document contents may involve certain patents,the identification of which is not the responsibility of the institution that releases the document.Main drafting organizations:Tsinghua University,Specialty Committee of Network Pharmacology of World Federation of Chinese Medicine Societies(WFCMS).
文摘The emerging development of connected and automated vehicles imposes a significant challenge on current vehicle control and transportation systems. This paper proposes a novel unified approach, Parallel Driving, a cloud-based cyberphysical-social systems(CPSS) framework aiming at synergizing connected automated driving. This study first introduces the CPSS and ACP-based intelligent machine systems. Then the parallel driving is proposed in the cyber-physical-social space,considering interactions among vehicles, human drivers, and information. Within the framework, parallel testing, parallel learning and parallel reinforcement learning are developed and concisely reviewed. Development on intelligent horizon(iHorizon)and its applications are also presented towards parallel horizon.The proposed parallel driving offers an ample solution for achieving a smooth, safe and efficient cooperation among connected automated vehicles with different levels of automation in future road transportation systems.
基金Project supported by the National Natural Science Foundation of China （Grant No 60404005）.
文摘In this paper we numerically investigate the chaotic behaviours of the fractional-order Ikeda delay system. The results show that chaos exists in the fractional-order Ikeda delay system with order less than 1. The lowest order for chaos to be a, ble to appear in this system is found to be 0.1. Master-slave synchronization of chaotic fractional-order Ikeda delay systems with linear coupling is also studied.
基金Supported by the National Natural Science Foundation of China(No.81225025 and 91229201)
文摘Traditional Chinese medicine（TCM） holds a holistic theory, and specializes in balancing disordered human body using numerous natural products, particularly Chinese herbal formulae. TCM has certain treatment advantages for patients suffering from various complex diseases. However, due to the complex nature of TCM, it remains difficult to unveil such holistic medicine by the current reductionism research strategies, which treat both herbal ingredients and targets in isolation. Recently, an emerging network pharmacology approach has been introduced to tackle this bottleneck problem. A TCM-derived novel therapeutic concept, ＂network target＂, which is different from the Western medicine＇s ＂onetarget＂ concept, has been proposed from China. The network target strategy is able to illustrate the complex interactions among the biological systems, drugs, and complex diseases from a network perspective, and thus provides an innovative approach to access ancient remedies in a precision manner and at a systematic level, which also highlights TCM＇s potential in current medical systems.
文摘Smartphones and mobile tablets are rapidly becoming indispensable in daily life. Android has been the most popular mobile operating system since 2012. However, owing to the open nature of Android, countless malwares are hidden in a large number of benign apps in Android markets that seriously threaten Android security. Deep learning is a new area of machine learning research that has gained increasing attention in artificial intelligence. In this study, we propose to associate the features from the static analysis with features from dynamic analysis of Android apps and characterize malware using deep learning techniques. We implement an online deep-learning-based Android malware detection engine（Droid Detector） that can automatically detect whether an app is a malware or not. With thousands of Android apps, we thoroughly test Droid Detector and perform an indepth analysis on the features that deep learning essentially exploits to characterize malware. The results show that deep learning is suitable for characterizing Android malware and especially effective with the availability of more training data. Droid Detector can achieve 96.76% detection accuracy, which outperforms traditional machine learning techniques. An evaluation of ten popular anti-virus softwares demonstrates the urgency of advancing our capabilities in Android malware detection.
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(61603169,61773192,61803192)in part by the funding from Shandong Provincial Key Laboratory for Novel Distributed Computer Software Technologyin part by Singapore National Research Foundation(NRF-RSS2016-004)
文摘Flexible job shop scheduling problems(FJSP)have received much attention from academia and industry for many years.Due to their exponential complexity,swarm intelligence(SI)and evolutionary algorithms(EA)are developed,employed and improved for solving them.More than 60%of the publications are related to SI and EA.This paper intents to give a comprehensive literature review of SI and EA for solving FJSP.First,the mathematical model of FJSP is presented and the constraints in applications are summarized.Then,the encoding and decoding strategies for connecting the problem and algorithms are reviewed.The strategies for initializing algorithms?population and local search operators for improving convergence performance are summarized.Next,one classical hybrid genetic algorithm(GA)and one newest imperialist competitive algorithm(ICA)with variables neighborhood search(VNS)for solving FJSP are presented.Finally,we summarize,discus and analyze the status of SI and EA for solving FJSP and give insight into future research directions.
基金co-supported in part by the National Natural Science Foundation of China (Nos. 61301205 and 61571160)the Natural Scientific Research Innovation Foundation at Harbin Institute of Technology (No. HIT.NSRIF.2014017)
文摘Lithium-ion batteries have become the third-generation space batteries and are widely utilized in a series of spacecraft. Remaining Useful Life （RUL） estimation is essential to a spacecraft as the battery is a critical part and determines the lifetime and reliability. The Relevance Vector Machine （RVM） is a data-driven algorithm used to estimate a battery＇s RUL due to its sparse feature and uncertainty management capability. Especially, some of the regressive cases indicate that the RVM can obtain a better short-term prediction performance rather than long-term prediction. As a nonlinear kernel learning algorithm, the coefficient matrix and relevance vectors are fixed once the RVM training is conducted. Moreover, the RVM can be simply influenced by the noise with the training data. Thus, this work proposes an iterative updated approach to improve the long-term prediction performance for a battery＇s RUL prediction. Firstly, when a new estimator is output by the RVM, the Kalman filter is applied to optimize this estimator with a physical degradation model. Then, this optimized estimator is added into the training set as an on-line sample, the RVM model is re-trained, and the coefficient matrix and relevance vectors can be dynamically adjusted to make next iterative prediction. Experimental results with a commercial battery test data set and a satellite battery data set both indicate that the proposed method can achieve a better performance for RUL estimation.
基金supported by the International Postdoctoral Exchange Fellowship Program(20180025)National Natural Science Foundation of China(51703180)+2 种基金China Postdoctoral Science Foundation(2018M630191,2017M610634)Shaanxi Postdoctoral Science Foundation(2017BSHEDZZ73)Fundamental Research Funds for the Central Universities(xpt012020006,xjj2017024).
文摘The application of intelligence to manufacturing has emerged as a compelling topic for researchers and industries around the world.However,different terminologies,namely smart manufacturing(SM)and intelligent manufacturing(IM),have been applied to what may be broadly characterized as a similar paradigm by some researchers and practitioners.While SM and IM are similar,they are not identical.From an evolutionary perspective,there has been little consideration on whether the definition,thought,connotation,and technical development of the concepts of SM or IM are consistent in the literature.To address this gap,the work performs a qualitative and quantitative investigation of research literature to systematically compare inherent differences of SM and IM and clarify the relationship between SM and IM.A bibliometric analysis of publication sources,annual publication numbers,keyword frequency,and top regions of research and development establishes the scope and trends of the currently presented research.Critical topics discussed include origin,definitions,evolutionary path,and key technologies of SM and IM.The implementation architecture,standards,and national focus are also discussed.In this work,a basis to understand SM and IM is provided,which is increasingly important because the trend to merge both terminologies rises in Industry 4.0 as intelligence is being rapidly applied to modern manufacturing and human–cyber–physical systems.
基金supported by the National Natural Science Foundation of China(Nos.61371128,U1611461,61425025,and 61532005)
文摘Cross-media analysis and reasoning is an active research area in computer science, and a promising direction for artificial intelligence. However, to the best of our knowledge, no existing work has summarized the state-of-the-art methods for cross-media analysis and reasoning or presented advances, challenges, and future directions for the field. To address these issues, we provide an overview as follows： （1） theory and model for cross-media uniform representation; （2） cross-media correlation understanding and deep mining; （3） cross-media knowledge graph construction and learning methodologies; （4） cross-media knowledge evolution and reasoning; （5） cross-media description and generation; （6） cross-media intelligent engines; and （7） cross-media intelligent applications. By presenting approaches, advances, and future directions in cross-media analysis and reasoning, our goal is not only to draw more attention to the state-of-the-art advances in the field, but also to provide technical insights by discussing the challenges and research directions in these areas.
基金Project (No. 60174009) supported by the National Natural ScienceFoundation of China
文摘Capacitated vehicle routing problem (CVRP) is an NP-hard problem. For large-scale problems, it is quite difficult to achieve an optimal solution with traditional optimization methods due to the high computational complexity. A new hybrid ap- proximation algorithm is developed in this work to solve the problem. In the hybrid algorithm, discrete particle swarm optimiza- tion (DPSO) combines global search and local search to search for the optimal results and simulated annealing (SA) uses certain probability to avoid being trapped in a local optimum. The computational study showed that the proposed algorithm is a feasible and effective approach for capacitated vehicle routing problem, especially for large scale problems.
基金National Natural Science Foundation of China (Grant Nos. 81630103, 91729301, 91229201 and 81225025)the Project of Tsinghua-Fuzhou Insititute for Data Technology (TFIDT2018001)to S. Li and grants 61673231 and 61721003 to X. Zhang.
文摘The development of gastritis is associated with an increased risk of gastric cancer. Current invasive gastritis diagnostic methods are not suitable for monitoring progressIn this work based on 78 gastritis patients and 50 healthy individuals, we observed that the variation of tongue-coating microbiota was associated with the occurrenee and development of gastritis. Twenty-one microbial species were identified for differentiating tongue-coating microbiomes of gastritis and healthy individuals. Pathways such as microbial metabolism in diverse environments, biosynthesis of antibiotics and bacterial chemotaxis were up-regulated in gastritis patients. The abundance of Campylobacter concisus was found associated with the gastric precancerous cascade. Furthermore, Campylobacter concisus could be detected in tongue coating and gastric fluid in a validation cohort containing 38 gastritis patients. These observations provided biological evidence of tongue diagnosis in traditional Chinese medicine, and indicated that tongue-coating microbiome could be a potential non-invasive biomarker, which might be suitable for long-term monitoring of gastritis.
基金The National Natural Science Foundation of China(contract grant number:81825012,81730048,81671126 to XL and contract grant number:81471390 to NM).
文摘Background and purpose We investigated the baseline demographics of patients with severe unilateral atherosclerotic stenosis of the middle cerebral artery(MCA)using multimodal MRI and evaluated the haemodynamic impairments and plaque characteristics of patients who had a recurrent stroke.Materials and methods We retrospectively recruited consecutive patients with severe unilateral atherosclerotic MCA stenosis who underwent arterial spin labelling(ASL)with postlabelling delay(PLD)of 1.5 and 2.5 s,and vessel wall MRI.For each PLD,cerebral blood flow(CBF)maps were generated.Hypoperfusion volume ratio(HVR)from 2 PLD CBF was calculated.An HVR value≥50%was considered as severe HVR.Plaque areas,plaque burden,plaque length and remodelling index were measured.Plaque enhancement at maximal lumen narrowing site were graded.Baseline clinical and imaging characteristics were compared between patients with(event+)and without(event?)1 year ischaemic events.Results Forty-three patients(47.23±12.15 years;28 men)were enrolled in this study.Seven patients had an HVR≥50%.During the 1-year follow-up,7 patients had experienced a recurrent stroke.HVR were significantly higher in the event+than event?(53.17%±29.82%vs 16.9%±15.57%,p=0.0002),whereas no significant difference was detected in plaque areas,plaque burden,remodelling index,plaque length and plaque enhancement grade.The multivariable analysis revealed that a severe HVR was significantly associated with a recurrent stroke(Odds ratio=12.93,95%confidence interval 1.57 to 106.24,p=0.017)after adjusted by hypertension and smoking.Conclusion HVR obtained from two PLD ASL may be a useful imaging predictor of recurrent stroke.
基金Project supported by National Natural Science Foundation of China （51177101）
文摘CF3I has been widely considered as a potential alternative for SF6,because it has low global warming potential(GWP)but an insulation capability that is 1.2 times greater than that of SF6.In this paper,the electron swarm parameters of CF3I and its gas mixture with N2,including the effective ionization coefficient and electron drift velocity,are examined theoretically through the Boltzmann equation method in the condition of steady-state Townsend(SST)experiments.Based on the derived data of the limiting field strength of CF3I-N2gas mixture,taking into consideration of environmental aspects such as GWP,ozone depletion potential(ODP),liquefaction temperature and toxicity,we studied the possibility of applying the gas mixture as the insulation medium in gas-insulated switchgears(GIS)or cubic type gas-insulated switchgears(C-GIS).It is found that CF3I-N2gas mixtures contained 30%～70%CF3I perform comprehensively better than pure SF6and compressed N2,and especially in medium and low voltage environments,the boiling point of CF3I-N2gas mixture meets the domestic and global requirements in mid-low latitude regions.Therefore we conclude that the gas mixture is acceptable for replacing SF6as the insulation medium in C-GISs.
