Nowadays,advances in communication technology and cloud computing have spawned a variety of smart mobile devices,which will generate a great amount of computing-intensive businesses,and require corresponding resources...Nowadays,advances in communication technology and cloud computing have spawned a variety of smart mobile devices,which will generate a great amount of computing-intensive businesses,and require corresponding resources of computation and communication.Multiaccess edge computing(MEC)can offload computing-intensive tasks to the nearby edge servers,which alleviates the pressure of devices.Ultra-dense network(UDN)can provide effective spectrum resources by deploying a large number of micro base stations.Furthermore,network slicing can support various applications in different communication scenarios.Therefore,this paper integrates the ultra-dense network slicing and the MEC technology,and introduces a hybrid computing offloading strategy in order to satisfy various quality of service(QoS)of edge devices.In order to dynamically allocate limited resources,the above problem is formulated as multiagent distributed deep reinforcement learning(DRL),which will achieve low overhead computation offloading strategy and real-time resource allocation decisions.In this context,federated learning is added to train DRL agents in a distributed manner,where each agent is dedicated to exploring actions composed of offloading decisions and allocating resources,so as to jointly optimize system delay and energy consumption.Simulation results show that the proposed learning algorithm has better performance compared with other strategies in literature.展开更多
目的:探究血清CC趋化因子配体2(CCL2)、血管内皮生长因子(VEGFA)与非酒精性脂肪性肝病(NAFLD)患者肝纤维化和代谢综合征(MS)的关系。方法:选取2022年1月至2024年1月在本院就诊的NAFLD患者116例(NAFLD组)作为研究对象,根据患者否并发MS分...目的:探究血清CC趋化因子配体2(CCL2)、血管内皮生长因子(VEGFA)与非酒精性脂肪性肝病(NAFLD)患者肝纤维化和代谢综合征(MS)的关系。方法:选取2022年1月至2024年1月在本院就诊的NAFLD患者116例(NAFLD组)作为研究对象,根据患者否并发MS分为MS组(42例)和非MS组(74例),另取同期体检健康者66例作为对照组。收集所有受试者的临床资料;采用ELISA法检测血清中CCL2和VEGFA表达量;Pearson法分析血清CCL2、VEGFA水平与肝纤维化及MS相关指标的相关性;Logistic多因素分析影响NAFLD患者并发MS的因素;受试者工作特征曲线分析血清CCL2和VEGFA水平对NAFLD患者并发MS的预测价值。结果:NAFLD患者血清CCL2、VEGFA水平以及肝纤维化指标层黏连蛋白(LN)、透明质酸(HA)、Ⅲ型前胶原(PCⅢ)、Ⅳ型胶原(ⅣC)均显著高于对照组(P<0.05)。MS组的收缩压、舒张压、空腹血糖(FPG)、餐后2 h血糖(2 h PG)、甘油三椡(TG)、CCL2、VEGFA水平显著高于非MS组,丙氨酸氨基转移酶(ALT)、天门冬氨酸氨基转移酶(AST)水平显著低于非MS组(P<0.05)。血清中CCL2和VEGFA水平与LN、HA、PCⅢ、ⅣC、收缩压、舒张压、FPG、2 h PG、TG呈正相关,与ALT、AST呈负相关(P<0.05)。收缩压、舒张压、FPG、2 h PG、TG、CCL2、VEGFA是影响NAFLD患者并发MS的危险因素,ALT、AST是影响NAFLD患者并发MS的保护因素(P<0.05)。血清CCL2和VEGFA水平以及联合预测NAFLD患者并发MS情况的曲线下面积分别为0.842、0.884和0.938,联合预测优于各自单独预测(Z_(联合-CCL2)=2.959、Z_(联合-VEGFA)=2.731,P=0.003、0.006)。结论:NAFLD患者血清CCL2和VEGFA水平升高,且二者与NAFLD患者肝纤维化和MS密切相关,二者联合对NAFLD患者并发MS具有较高的预测价值。展开更多
目的:探讨基于冠状动脉CT血管成像(CCTA)的冠状动脉周围脂肪衰减指数(FAI)与冠状动脉斑块定量参数的相关性。方法:前瞻性连续纳入因怀疑冠心病接受CCTA检查的250例患者,分析750支冠状动脉主干[左前降支(LAD)、左回旋支(LCX)、右冠状动脉...目的:探讨基于冠状动脉CT血管成像(CCTA)的冠状动脉周围脂肪衰减指数(FAI)与冠状动脉斑块定量参数的相关性。方法:前瞻性连续纳入因怀疑冠心病接受CCTA检查的250例患者,分析750支冠状动脉主干[左前降支(LAD)、左回旋支(LCX)、右冠状动脉(RCA)]。采用RevolutionCT Victor 256进行CCTA扫描,通过联影uAI平台获取冠状动脉疾病报告和数据系统(CAD-RADS)评分、节段狭窄评分(SSS)、节段累及评分(SIS),总斑块长度(TPL)、总斑块体积(TPV)、钙化斑块体积(CPV)、非钙化斑块体积(NCPV)、脂质斑块体积(LPV)及总斑块负荷(TPB%)等斑块定量参数,同时测量三大主干的FAI值,定义FAI≥-70.1HU为阳性组。结果:115例患者的175支冠脉为FAI阳性组,两组间比较分析结果显示,FAI阳性组的CAD-RADS评分、SSS、SIS及TPL、TPV、CPV、NCPV、LPV、TPB%均显著高于FAI阴性组(P<0.01)。进一步的二元logistic回归分析显示,CAD-RADS评分(OR=1.79,P<0.01)和LPV(OR=1.03,P=0.01)是FAI阳性的独立危险因素:受试者工作特征(ROC)曲线分析显示,二者联合的ROC曲线下面积(AUC)=0.79。结论:冠周脂肪FAI与冠脉狭窄程度及LPV呈正相关,验证了冠周炎症与粥样硬化及斑块脂质成分的关联。展开更多
In this study,we investigate the ef-ficacy of a hybrid parallel algo-rithm aiming at enhancing the speed of evaluation of two-electron repulsion integrals(ERI)and Fock matrix generation on the Hygon C86/DCU(deep compu...In this study,we investigate the ef-ficacy of a hybrid parallel algo-rithm aiming at enhancing the speed of evaluation of two-electron repulsion integrals(ERI)and Fock matrix generation on the Hygon C86/DCU(deep computing unit)heterogeneous computing platform.Multiple hybrid parallel schemes are assessed using a range of model systems,including those with up to 1200 atoms and 10000 basis func-tions.The findings of our research reveal that,during Hartree-Fock(HF)calculations,a single DCU ex-hibits 33.6 speedups over 32 C86 CPU cores.Compared with the efficiency of Wuhan Electronic Structure Package on Intel X86 and NVIDIA A100 computing platform,the Hygon platform exhibits good cost-effective-ness,showing great potential in quantum chemistry calculation and other high-performance scientific computations.展开更多
Federated learning(FL)is a distributed machine learning paradigm for edge cloud computing.FL can facilitate data-driven decision-making in tactical scenarios,effectively addressing both data volume and infrastructure ...Federated learning(FL)is a distributed machine learning paradigm for edge cloud computing.FL can facilitate data-driven decision-making in tactical scenarios,effectively addressing both data volume and infrastructure challenges in edge environments.However,the diversity of clients in edge cloud computing presents significant challenges for FL.Personalized federated learning(pFL)received considerable attention in recent years.One example of pFL involves exploiting the global and local information in the local model.Current pFL algorithms experience limitations such as slow convergence speed,catastrophic forgetting,and poor performance in complex tasks,which still have significant shortcomings compared to the centralized learning.To achieve high pFL performance,we propose FedCLCC:Federated Contrastive Learning and Conditional Computing.The core of FedCLCC is the use of contrastive learning and conditional computing.Contrastive learning determines the feature representation similarity to adjust the local model.Conditional computing separates the global and local information and feeds it to their corresponding heads for global and local handling.Our comprehensive experiments demonstrate that FedCLCC outperforms other state-of-the-art FL algorithms.展开更多
Graph computing has become pervasive in many applications due to its capacity to represent complex relationships among different objects in the big data era.However,general-purpose architectures are computationally in...Graph computing has become pervasive in many applications due to its capacity to represent complex relationships among different objects in the big data era.However,general-purpose architectures are computationally inefficient for graph algorithms,and dedicated architectures can provide high efficiency,but lack flexibility.To address these challenges,this paper proposes ParaGraph,a reduced instruction set computing-five(RISC-V)-based software-hardware co-designed graph computing accelerator that can process graph algorithms in parallel,and also establishes a performance evaluation model to assess the efficiency of co-acceleration.ParaGraph handles parallel processing of typical graph algorithms on the hardware side,while performing overall functional control on the software side with custom designed instructions.ParaGraph is verified on the XCVU440 field-programmable gate array(FPGA)board with E203,a RISC-V processor.Compared with current mainstream graph computing accelerators,ParaGraph consumes 7.94%less block RAM(BRAM)resources than ThunderGP.Its power consumption is reduced by 86.90%,24.90%,and 76.38%compared with ThunderGP,HitGraph,and GraphS,respectively.The power efficiency of connected components(CC)and degree centrality(DC)algorithms is improved by an average of 6.50 times over ThunderGP,2.51 times over HitGraph,and 3.99 times over GraphS.The software-hardware co-design acceleration performance indicators H/W.Cap for CC and DC are 13.02 and 14.02,respectively.展开更多
The increasing popularity of quantum computing has resulted in a considerable rise in demand for cloud quantum computing usage in recent years.Nevertheless,the rapid surge in demand for cloud-based quantum computing r...The increasing popularity of quantum computing has resulted in a considerable rise in demand for cloud quantum computing usage in recent years.Nevertheless,the rapid surge in demand for cloud-based quantum computing resources has led to a scarcity.In order to meet the needs of an increasing number of researchers,it is imperative to facilitate efficient and flexible access to computing resources in a cloud environment.In this paper,we propose a novel quantum computing paradigm,Virtual QPU(VQPU),which addresses this issue and enhances quantum cloud throughput with guaranteed circuit fidelity.The proposal introduces three innovative concepts:(1)The integration of virtualization technology into the field of quantum computing to enhance quantum cloud throughput.(2)The introduction of an asynchronous execution of circuits methodology to improve quantum computing flexibility.(3)The development of a virtual QPU allocation scheme for quantum tasks in a cloud environment to improve circuit fidelity.The concepts have been validated through the utilization of a self-built simulated quantum cloud platform.展开更多
Organic electrochemical transistor(OECT)devices demonstrate great promising potential for reservoir computing(RC)systems,but their lack of tunable dynamic characteristics limits their application in multi-temporal sca...Organic electrochemical transistor(OECT)devices demonstrate great promising potential for reservoir computing(RC)systems,but their lack of tunable dynamic characteristics limits their application in multi-temporal scale tasks.In this study,we report an OECT-based neuromorphic device with tunable relaxation time(τ)by introducing an additional vertical back-gate electrode into a planar structure.The dual-gate design enablesτreconfiguration from 93 to 541 ms.The tunable relaxation behaviors can be attributed to the combined effects of planar-gate induced electrochemical doping and back-gateinduced electrostatic coupling,as verified by electrochemical impedance spectroscopy analysis.Furthermore,we used theτ-tunable OECT devices as physical reservoirs in the RC system for intelligent driving trajectory prediction,achieving a significant improvement in prediction accuracy from below 69%to 99%.The results demonstrate that theτ-tunable OECT shows a promising candidate for multi-temporal scale neuromorphic computing applications.展开更多
文摘Nowadays,advances in communication technology and cloud computing have spawned a variety of smart mobile devices,which will generate a great amount of computing-intensive businesses,and require corresponding resources of computation and communication.Multiaccess edge computing(MEC)can offload computing-intensive tasks to the nearby edge servers,which alleviates the pressure of devices.Ultra-dense network(UDN)can provide effective spectrum resources by deploying a large number of micro base stations.Furthermore,network slicing can support various applications in different communication scenarios.Therefore,this paper integrates the ultra-dense network slicing and the MEC technology,and introduces a hybrid computing offloading strategy in order to satisfy various quality of service(QoS)of edge devices.In order to dynamically allocate limited resources,the above problem is formulated as multiagent distributed deep reinforcement learning(DRL),which will achieve low overhead computation offloading strategy and real-time resource allocation decisions.In this context,federated learning is added to train DRL agents in a distributed manner,where each agent is dedicated to exploring actions composed of offloading decisions and allocating resources,so as to jointly optimize system delay and energy consumption.Simulation results show that the proposed learning algorithm has better performance compared with other strategies in literature.
文摘目的:探究血清CC趋化因子配体2(CCL2)、血管内皮生长因子(VEGFA)与非酒精性脂肪性肝病(NAFLD)患者肝纤维化和代谢综合征(MS)的关系。方法:选取2022年1月至2024年1月在本院就诊的NAFLD患者116例(NAFLD组)作为研究对象,根据患者否并发MS分为MS组(42例)和非MS组(74例),另取同期体检健康者66例作为对照组。收集所有受试者的临床资料;采用ELISA法检测血清中CCL2和VEGFA表达量;Pearson法分析血清CCL2、VEGFA水平与肝纤维化及MS相关指标的相关性;Logistic多因素分析影响NAFLD患者并发MS的因素;受试者工作特征曲线分析血清CCL2和VEGFA水平对NAFLD患者并发MS的预测价值。结果:NAFLD患者血清CCL2、VEGFA水平以及肝纤维化指标层黏连蛋白(LN)、透明质酸(HA)、Ⅲ型前胶原(PCⅢ)、Ⅳ型胶原(ⅣC)均显著高于对照组(P<0.05)。MS组的收缩压、舒张压、空腹血糖(FPG)、餐后2 h血糖(2 h PG)、甘油三椡(TG)、CCL2、VEGFA水平显著高于非MS组,丙氨酸氨基转移酶(ALT)、天门冬氨酸氨基转移酶(AST)水平显著低于非MS组(P<0.05)。血清中CCL2和VEGFA水平与LN、HA、PCⅢ、ⅣC、收缩压、舒张压、FPG、2 h PG、TG呈正相关,与ALT、AST呈负相关(P<0.05)。收缩压、舒张压、FPG、2 h PG、TG、CCL2、VEGFA是影响NAFLD患者并发MS的危险因素,ALT、AST是影响NAFLD患者并发MS的保护因素(P<0.05)。血清CCL2和VEGFA水平以及联合预测NAFLD患者并发MS情况的曲线下面积分别为0.842、0.884和0.938,联合预测优于各自单独预测(Z_(联合-CCL2)=2.959、Z_(联合-VEGFA)=2.731,P=0.003、0.006)。结论:NAFLD患者血清CCL2和VEGFA水平升高,且二者与NAFLD患者肝纤维化和MS密切相关,二者联合对NAFLD患者并发MS具有较高的预测价值。
文摘目的:探讨基于冠状动脉CT血管成像(CCTA)的冠状动脉周围脂肪衰减指数(FAI)与冠状动脉斑块定量参数的相关性。方法:前瞻性连续纳入因怀疑冠心病接受CCTA检查的250例患者,分析750支冠状动脉主干[左前降支(LAD)、左回旋支(LCX)、右冠状动脉(RCA)]。采用RevolutionCT Victor 256进行CCTA扫描,通过联影uAI平台获取冠状动脉疾病报告和数据系统(CAD-RADS)评分、节段狭窄评分(SSS)、节段累及评分(SIS),总斑块长度(TPL)、总斑块体积(TPV)、钙化斑块体积(CPV)、非钙化斑块体积(NCPV)、脂质斑块体积(LPV)及总斑块负荷(TPB%)等斑块定量参数,同时测量三大主干的FAI值,定义FAI≥-70.1HU为阳性组。结果:115例患者的175支冠脉为FAI阳性组,两组间比较分析结果显示,FAI阳性组的CAD-RADS评分、SSS、SIS及TPL、TPV、CPV、NCPV、LPV、TPB%均显著高于FAI阴性组(P<0.01)。进一步的二元logistic回归分析显示,CAD-RADS评分(OR=1.79,P<0.01)和LPV(OR=1.03,P=0.01)是FAI阳性的独立危险因素:受试者工作特征(ROC)曲线分析显示,二者联合的ROC曲线下面积(AUC)=0.79。结论:冠周脂肪FAI与冠脉狭窄程度及LPV呈正相关,验证了冠周炎症与粥样硬化及斑块脂质成分的关联。
基金supported by the National Natural Science Foundation of China(No.22373112 to Ji Qi,No.22373111 and 21921004 to Minghui Yang)GH-fund A(No.202107011790)。
文摘In this study,we investigate the ef-ficacy of a hybrid parallel algo-rithm aiming at enhancing the speed of evaluation of two-electron repulsion integrals(ERI)and Fock matrix generation on the Hygon C86/DCU(deep computing unit)heterogeneous computing platform.Multiple hybrid parallel schemes are assessed using a range of model systems,including those with up to 1200 atoms and 10000 basis func-tions.The findings of our research reveal that,during Hartree-Fock(HF)calculations,a single DCU ex-hibits 33.6 speedups over 32 C86 CPU cores.Compared with the efficiency of Wuhan Electronic Structure Package on Intel X86 and NVIDIA A100 computing platform,the Hygon platform exhibits good cost-effective-ness,showing great potential in quantum chemistry calculation and other high-performance scientific computations.
基金supported by the Natural Science Foundation of Xinjiang Uygur Autonomous Region(Grant No.2022D01B 187)。
文摘Federated learning(FL)is a distributed machine learning paradigm for edge cloud computing.FL can facilitate data-driven decision-making in tactical scenarios,effectively addressing both data volume and infrastructure challenges in edge environments.However,the diversity of clients in edge cloud computing presents significant challenges for FL.Personalized federated learning(pFL)received considerable attention in recent years.One example of pFL involves exploiting the global and local information in the local model.Current pFL algorithms experience limitations such as slow convergence speed,catastrophic forgetting,and poor performance in complex tasks,which still have significant shortcomings compared to the centralized learning.To achieve high pFL performance,we propose FedCLCC:Federated Contrastive Learning and Conditional Computing.The core of FedCLCC is the use of contrastive learning and conditional computing.Contrastive learning determines the feature representation similarity to adjust the local model.Conditional computing separates the global and local information and feeds it to their corresponding heads for global and local handling.Our comprehensive experiments demonstrate that FedCLCC outperforms other state-of-the-art FL algorithms.
基金Supported by the National Key R&D Program of China(No.2022ZD0119001)the National Natural Science Foundation of China(No.61834005)+1 种基金the Shaanxi Province Key R&D Plan(No.2022GY-027,2021GY-029)the Key Scientific Research Project of Shaanxi Department of Education(No.22JY060).
文摘Graph computing has become pervasive in many applications due to its capacity to represent complex relationships among different objects in the big data era.However,general-purpose architectures are computationally inefficient for graph algorithms,and dedicated architectures can provide high efficiency,but lack flexibility.To address these challenges,this paper proposes ParaGraph,a reduced instruction set computing-five(RISC-V)-based software-hardware co-designed graph computing accelerator that can process graph algorithms in parallel,and also establishes a performance evaluation model to assess the efficiency of co-acceleration.ParaGraph handles parallel processing of typical graph algorithms on the hardware side,while performing overall functional control on the software side with custom designed instructions.ParaGraph is verified on the XCVU440 field-programmable gate array(FPGA)board with E203,a RISC-V processor.Compared with current mainstream graph computing accelerators,ParaGraph consumes 7.94%less block RAM(BRAM)resources than ThunderGP.Its power consumption is reduced by 86.90%,24.90%,and 76.38%compared with ThunderGP,HitGraph,and GraphS,respectively.The power efficiency of connected components(CC)and degree centrality(DC)algorithms is improved by an average of 6.50 times over ThunderGP,2.51 times over HitGraph,and 3.99 times over GraphS.The software-hardware co-design acceleration performance indicators H/W.Cap for CC and DC are 13.02 and 14.02,respectively.
文摘The increasing popularity of quantum computing has resulted in a considerable rise in demand for cloud quantum computing usage in recent years.Nevertheless,the rapid surge in demand for cloud-based quantum computing resources has led to a scarcity.In order to meet the needs of an increasing number of researchers,it is imperative to facilitate efficient and flexible access to computing resources in a cloud environment.In this paper,we propose a novel quantum computing paradigm,Virtual QPU(VQPU),which addresses this issue and enhances quantum cloud throughput with guaranteed circuit fidelity.The proposal introduces three innovative concepts:(1)The integration of virtualization technology into the field of quantum computing to enhance quantum cloud throughput.(2)The introduction of an asynchronous execution of circuits methodology to improve quantum computing flexibility.(3)The development of a virtual QPU allocation scheme for quantum tasks in a cloud environment to improve circuit fidelity.The concepts have been validated through the utilization of a self-built simulated quantum cloud platform.
基金supported by the National Key Research and Development Program of China under Grant 2022YFB3608300in part by the National Nature Science Foundation of China(NSFC)under Grants 62404050,U2341218,62574056,62204052。
文摘Organic electrochemical transistor(OECT)devices demonstrate great promising potential for reservoir computing(RC)systems,but their lack of tunable dynamic characteristics limits their application in multi-temporal scale tasks.In this study,we report an OECT-based neuromorphic device with tunable relaxation time(τ)by introducing an additional vertical back-gate electrode into a planar structure.The dual-gate design enablesτreconfiguration from 93 to 541 ms.The tunable relaxation behaviors can be attributed to the combined effects of planar-gate induced electrochemical doping and back-gateinduced electrostatic coupling,as verified by electrochemical impedance spectroscopy analysis.Furthermore,we used theτ-tunable OECT devices as physical reservoirs in the RC system for intelligent driving trajectory prediction,achieving a significant improvement in prediction accuracy from below 69%to 99%.The results demonstrate that theτ-tunable OECT shows a promising candidate for multi-temporal scale neuromorphic computing applications.
