检索结果-南通市图书馆

维普期刊数据库

Learning in games:a systematic review

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第7期67卷 133-159页

作者： Rong-Jun QIN yang yu National Key Laboratory for Novel Software Technology Nanjing University Polixir Technologies

Game theory studies the mathematical models for self-interested individuals. Nash equilibrium is arguably the most central solution in game theory. While finding the Nash equilibrium in general is known as polynomial parity arguments on directed graphs(PPAD)-complete, learning in games provides an alternative to approximate Nash equilibrium, which iteratively updates the player's strategy through interactions with other players. Rules and models have been developed for learning in games, such as fictitious play and no-regret learning. Particularly, with recent advances in online learning and deep reinforcement learning,techniques from these fields greatly boost the breakthroughs in learning in games from theory to *** a result, we have witnessed many superhuman game AI systems. The techniques used in these systems evolve from conventional search and learning to purely reinforcement learning(RL)-style learning methods,gradually getting rid of the domain knowledge. In this article, we systematically review the above techniques,discuss the trend of basic learning rules towards a unified framework, and recap applications in large ***, we discuss some future directions and make the prospect of future game AI systems. We hope this article will give some insights into designing novel approaches.

关键词： non-cooperative games learning in games no-regret learning reinforcement learning superhuman AI

Evolution of inner wall wrinkle defects in the sinking zone of a thick-walled steel tube during radial forging

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Advances in Manufacturing 2024年第2期12卷 396-408页

作者： yu-Zhao yang Cheng Xu Li-Xia Fan School of Mechanical Engineering Nanjing University of Science and TechnologyNanjing210094People’s Republic of China

An axial wrinkle defect was observed in the inner wall of the sinking zone of a thick-wall steel tube processed by cold radial *** can evolve into *** present study focuses on the evolution of wrinkles and the effects of process parameters on them using a three-dimensional radial forging process finite element model,radial forging experiments,and surface morphology *** results indicated that the vertical section angle of the hammer die and the size of the tube blank significantly affect the evolution of *** height-to-width ratioλwas introduced to describe the morphology of *** had an approximately linear relationship with the radius reduction in the sinking *** was a linear correlation between the growth slope ofλand the axial to circumferential strain ratio|εr/εθ|,which can predict theλunder few process *** the 30SiMn2MoVA steel,at the junction of the forging and sinking zones,the threshold ofλof the wrinkle that can evolve into a fissure is approximately 1.123.

关键词： Radial forging Inner wall Wrinkle defects Evolution Sinking zone

Interface failure of segmental tunnel lining strengthened with steel plates based on fracture mechanics

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Structural and Civil Engineering 2024年第1期18卷 137-149页

作者： Yazhen SUN yang yu Jinchang WANG Longyan WANG School of Transportation and Geomatics Engineering Shenyang Jianzhu UniversityShenyang 110168China School of Civil Engineering Shenyang Jianzhu UniversityShenyang 110168China Institute of Transportation Engineering Zhejiang UniversityHangzhou 310058China Center for Balance Architecture Zhejiang UniversityHangzhou310058China

Segmental tunnel lining strengthened with steel plates is widely used worldwide to provide a permanent strengthening *** existing studies assume an ideal steel-concrete interface,ignoring discontinuous deformation characteristics,making it difficult to accurately analyze the strengthened structure’s failure *** this study,interfacial fracture mechanics of composite material was applied to the segmental tunnel lining strengthened with steel plates,and a numerical three-dimensional solid nonlinear model of the lining structure was established,combining the extended finite element method with a cohesive-zone model to account for the discontinuous deformation characteristics of the *** results accurately describe the crack propagation process,and are verified by full-scale ***,dynamic simulations based on the calibrated model were conducted to analyze the sliding failure and cracking of the steel-concrete ***,detailed location of the interface bonding failure are further verified by model *** results show that,the cracking failure and bond failure of the interface are the decisive factors determining the instability and failure of the strengthened *** proposed numerical analysis is a major step forward in revealing the interface failure mechanism of strengthened composite material structures.

关键词： segmental tunnel lining steel plate strengthening connecting interface cohesive-zone model extended finite element method

Preparation of maximally-entangled states with multiple cat-state qutrits in circuit QED

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of physics 2024年第3期19卷 239-251页

作者： Chui-Ping yang Jia-Heng Ni Liang Bin yu Zhang yang yu Qi-Ping Su Department of Physics Hangzhou Normal UniversityHangzhou 311121China School of Physics Nanjing UniversityNanjing 210093China Institute for Quantum Science and Technology University of CalgaryAlberta T2N 1N4Canada

In recent years,cat-state encoding and high-dimensional entanglement have attracted much ***,previous works are limited to generation of entangled states of cat-state qubits(two-dimensional entanglement with cat-state encoding),while how to prepare entangled states of cat-state qutrits or qudits(high-dimensional entanglement with cat-state encoding)has not been *** here propose to generate a maximally-entangled state of multiple cat-state qutrits(three-dimensional entanglement by cat-state encoding)in circuit *** entangled state is prepared with multiple microwave cavities coupled to a superconducting flux ququart(a four-level quantum system).This proposal operates essentially by the cavity-qutrit dispersive *** circuit hardware resource is minimized because only a coupler ququart is *** higher intermediate level of the ququart is occupied only for a short time,thereby decoherence from this level is greatly suppressed during the state ***,the state preparation time does not depend on the number of the qutrits,thus it does not increase with the number of the *** an example,our numerical simulations demonstrate that,with the present circuit QED technology,the high-fidelity preparation is feasible for a maximally-entangled state of two cat-state ***,we numerically analyze the effect of the inter-cavity crosstalk on the scalability of this *** proposal is universal and can be extended to accomplish the same task with multiple microwave or optical cavities coupled to a natural or artificial four-level atom.

关键词： maximally-entangled states cat state qutrit circuit QED

Stabilization and detection of labile reaction intermediates in supramolecular containers

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Chemical Letters 2024年第6期35卷 47-52页

作者： Rui Wang yang Liang Julius Rebek Jr. yang yu Center for Supramolecular Chemistry&Catalysis and Department of Chemistry College of ScienceShanghai UniversityShanghai 200444China Skaggs Institute for Chemical Biology and Department of Chemistry The Scripps Research InstituteLa JollaCA 92037United States

Detection and observation of reactive intermediates is an essential step in investigation of reaction ***,most reactive intermediates are unstable and present at low concentrations;their short lifetimes make them difficult to detect and *** containers offer opportunities for the stabilization and characterization of those labile species,through isolation from the media and protection inside the cavity of the *** this review,we summarize the examples of labile reaction intermediates that are stabilized and characterized with the help of supramolecular *** container compounds include carcerands,deep cavitands and amide *** focus on unstable guest species-cyclobutadiene,benzocyclopropenone,o-benzyne,1,2,4,6-cycloheptatetraene,anti-Bredt's olefin,fluorophenoxycarbene,O-acylisoamide,and hemiaminalthat act as intermediates in certain organic reactions.

关键词： Supramolecular containers Reaction intermediates Labile species stabilization Short-lived species detection Supramolecular chemistry

维普期刊数据库

A novel and promising engineering application of carbon dots:Enhancing the chloride binding performance of cement

在线全文

学校读者我要写书评

暂无评论

Chinese Chemical Letters 2024年第6期35卷 409-416页

作者： Wu-Jian Long yang yu Chuang He Guangdong Provincial Key Laboratory of Durability for Marine Civil Engineering Key Lab of Coastal Urban Resilient InfrastructureMOECollege of Civil and Transportation EngineeringShenzhen UniversityShenzhen 518060China

Corrosion of reinforcement induced by chloride invasion is extensively considered as the dominating deterioration mechanism of reinforced concrete(RC)structures,leading to serious safety hazards and tremendous economic ***,it still lacks well dispersive and cost-efficient nanomaterials to improve the anti-chloride-corrosion ability of RC ***,specific carbon dots(CDs)with high dispersity and low cost are deliberately designed,successfully prepared by hydrothermal processing,and then firstly applied to immensely enhance chloride binding performance of cement,thereby contributing to suppressing the corrosion of ***,the tailored CDs are composed of the carbon core with highly crystalline sp^(2)C structures and oxygen-containing groups connecting on the carbon core;The typical equilibrium test confirms that with respect to that of the blank cement paste,the chloride binding capacity of cement paste involving 0.2 wt%(by weight of cement)CDs is increased by 109% after 14-day exposure to 3 mol/L Na Cl solution;according to comprehensive analyses of phase compositions,the chloride binding mechanism of CDs-modified cement is rationally attributed to the fact that the incorporation of CDs advances the formation of calcium silicate hydrate(C-S-H)gels and Friedel's salt(Fs),thus enormously enhancing the physically adsorbed and chemically bound chloride ions of cement *** work not only firstly provides a novel high-dispersity and low-cost nanomaterial toward the durability enhancement of RC structures,but also broadens the application of CDs in the field of engineering,conducing to stimulating their industrialization development.

关键词： Carbon dots Chloride binding Cement Chloride binding mechanism Friedel’s salt

维普期刊数据库

MUS Model:A Deep Learning-Based Architecture for IoT Intrusion Detection

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第7期80卷 875-896页

作者： yu Yan yu yang Shen Fang Minna Gao Yiding Chen College of Information Engineering University of Engineering of the Chinese People’s Armed Police Force(PAP)Xi’an710000China College of Missile Engineering Rocket Force Engineering UniversityXi’an710000China

In the face of the effective popularity of the Internet of Things(IoT),but the frequent occurrence of cybersecurity incidents,various cybersecurity protection means have been proposed and *** them,Intrusion Detection System(IDS)has been proven to be stable and ***,traditional intrusion detection methods have shortcomings such as lowdetection accuracy and inability to effectively identifymalicious *** address the above problems,this paper fully considers the superiority of deep learning models in processing highdimensional data,and reasonable data type conversion methods can extract deep features and detect classification using advanced computer vision techniques to improve classification *** TransformField(MTF)method is used to convert 1Dnetwork traffic data into 2D images,and then the converted 2D images are filtered by UnsharpMasking to enhance the image details by sharpening;to further improve the accuracy of data classification and detection,unlike using the existing high-performance baseline image classification models,a soft-voting integrated model,which integrates three deep learning models,MobileNet,VGGNet and ResNet,to finally obtain an effective IoT intrusion detection architecture:the MUS *** types of experiments are conducted on the publicly available intrusion detection dataset CICIDS2018 and the IoT network traffic dataset N_BaIoT,and the results demonstrate that the accuracy of attack traffic detection is greatly improved,which is not only applicable to the IoT intrusion detection environment,but also to different types of attacks and different network environments,which confirms the effectiveness of the work done.

关键词： Cyberspace security intrusion detection deep learning Markov Transition Fields(MTF) soft voting integration

State and parameter identification of linearized water wave equation via adjoint method

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第10期67卷 283-297页

作者： yang yu Cheng-Zhong XU Hai-Long PEI Jinpeng yu School of Automation Qingdao University LAGEP Batiment CPE University Claude Bernard-Lyon 1 School of Automation Science and Engineering Key Laboratory of Autonomous Systems and Networked ControlMinistry of Education Guangdong Engineering Technology Research Center of Unmanned Aerial Vehicle SystemSouth China University of Technology

In this paper, we focus on the state and parameter identification problem of a hydrodynamical system. This system is modeled as a linearized water wave equation(LWWE), a hyperbolic state-space model coupled with a Laplace equation. We assume that the wave elevation at two distinct points is the only measurement of water waves. We show that the state and water depth can be reconstructed from this point measurement records. The identification problem is recast as an optimization problem over an infinite-dimensional space. We propose the adjoint method-based identification algorithm to generate an estimated state and water depth. We then performed a numerical simulation to show the effectiveness of our designed algorithm by comparing it with existing studies.

关键词： linearized water wave equation (LWWE) point observation state estimation parameter identification adjoint method numerical simulation

同方期刊数据库

Reinforcement Learning Based Quantization Strategy Optimal Assignment Algorithm for Mixed Precision

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第4期79卷 819-836页

作者： yuejiao Wang Zhong Ma Chaojie yang yu yang Lu Wei Xi’an Microelectronics Technology Institute Xi’an710065China

The quantization algorithm compresses the original network by reducing the numerical bit width of the model,which improves the computation speed. Because different layers have different redundancy and sensitivity to databit width. Reducing the data bit width will result in a loss of accuracy. Therefore, it is difficult to determinethe optimal bit width for different parts of the network with guaranteed accuracy. Mixed precision quantizationcan effectively reduce the amount of computation while keeping the model accuracy basically unchanged. In thispaper, a hardware-aware mixed precision quantization strategy optimal assignment algorithm adapted to low bitwidth is proposed, and reinforcement learning is used to automatically predict the mixed precision that meets theconstraints of hardware resources. In the state-space design, the standard deviation of weights is used to measurethe distribution difference of data, the execution speed feedback of simulated neural network accelerator inferenceis used as the environment to limit the action space of the agent, and the accuracy of the quantization model afterretraining is used as the reward function to guide the agent to carry out deep reinforcement learning training. Theexperimental results show that the proposed method obtains a suitable model layer-by-layer quantization strategyunder the condition that the computational resources are satisfied, and themodel accuracy is effectively *** proposed method has strong intelligence and certain universality and has strong application potential in thefield of mixed precision quantization and embedded neural network model deployment.

关键词： Mixed precision quantization quantization strategy optimal assignment reinforcement learning neural network model deployment