检索结果-南通市图书馆

维普期刊数据库

Robust cooperative multi-agent reinforcement learning via multi-view message certification

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第4期67卷 131-145页

作者： Lei YUAN tao jiang Lihe LI Feng CHEN Zongzhang ZHANG Yang YU National Key Laboratory for Novel Software Technology Nanjing University Polixir Technologies

Many multi-agent scenarios require message sharing among agents to promote coordination,hastening the robustness of multi-agent communication when policies are deployed in a message perturbation environment. Major relevant studies tackle this issue under specific assumptions, like a limited number of message channels would sustain perturbations, limiting the efficiency in complex scenarios. In this paper,we take a further step in addressing this issue by learning a robust cooperative multi-agent reinforcement learning via multi-view message certification, dubbed CroMAC. Agents trained under CroMAC can obtain guaranteed lower bounds on state-action values to identify and choose the optimal action under a worst-case deviation when the received messages are perturbed. Concretely, we first model multi-agent communication as a multi-view problem, where every message stands for a view of the state. Then we extract a certificated joint message representation by a multi-view variational autoencoder(MVAE) that uses a product-of-experts inference network. For the optimization phase, we do perturbations in the latent space of the state for a certificate guarantee. Then the learned joint message representation is used to approximate the certificated state representation during training. Extensive experiments in several cooperative multi-agent benchmarks validate the effectiveness of the proposed CroMAC.

关键词： multi-agent reinforcement learning robust communication adversarial training multi-view learning message certification

Unlocking supercapacitive energy storage potential:Catalyzing electrochemically inactive manganese oxides to active MnO_(2)via heterostructure reconstruction

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Nano Research 2024年第7期17卷 5897-5906页

作者： Baohong Zhang tao jiang Xinyan Zhou Xiaoyu Fan Binbin Jia Lidong Li School of Chemistry Beihang UniversityBeijing 100191China College of Materials and Chemical Engineering Key Laboratory of Inorganic Nonmetallic Crystalline and Energy Conversion MaterialsChinaThree Gorges UniversityYichang 443002China Chinese Academy of Sciences(CAS)Key Laboratory of Nanosystem and Hierarchy Fabrication CAS Center for Excellence in NanoscienceNational Center for Nanoscience and TechnologyBeijing 100190China Shanghai GTX Semiconductor Co. Ltd.Shanghai 201806China

Advancing supercapacitor system performance hinges on the innovation of novel electrode materials seamlessly integrated within distinct ***,we introduce a direct approach for crafting nanorod arrays featuring crystalline/amorphous CuO/MnO_(2)−*** reconfigured heterostructure results in an elevated content of electrochemically active MnO_(2).The nanorod arrays serve as efficient capacitive anodes and are easily prepared via low-potential electrochemical *** resulting structure spontaneously forms a p–n heterojunction,developing a built-in electric field that dramatically facilitates the charge transport *** intrinsic electric field,in conjunction with the crystalline/amorphous architecture,enables a large capacitance of 1.0 F·cm^(−2)at 1.0 mA·cm^(−2),an ultrahigh rate capability of approximately 85.4%at 15 mA·cm^(−2),and stable cycling performance with 92.4%retention after 10,000 *** calculations reveal that the presence of heterojunctions allows for the optimization of the electronic structure of this composite,leading to improved conductivity and optimized OH−adsorption *** work provides new insights into the rational design of heterogeneous nanostructures,which hold great potential in energy storage applications.

关键词： reconfiguration active MnO_(2) crystalline/amorphous structure p-n heterojunction supercapacitor

An Improved Solov2 Based on Attention Mechanism and Weighted Loss Function for Electrical Equipment Instance Segmentation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第1期78卷 677-694页

作者： Junpeng Wu Zhenpeng Liu Xingfan jiang Xinguang tao Ye Zhang Key Laboratory of Modern Power System Simulation and Control&Renewable Energy Technology Ministry of EducationNortheast Electric Power UniversityJilin132012China School of Electrical Engineering Northeast Electric Power UniversityJilin132012China Baishan Power Supply Company State Grid Jilin Electric Power CoLtd.Jilin134300China

The current existing problem of deep learning framework for the detection and segmentation of electrical equipment is dominantly related to low *** of the reliable,safe and easy-to-operate technology provided by deep learning-based video surveillance for unmanned inspection of electrical equipment,this paper uses the bottleneck attention module(BAM)attention mechanism to improve the Solov2 model and proposes a new electrical equipment segmentation ***,the BAM attention mechanism is integrated into the feature extraction network to adaptively learn the correlation between feature channels,thereby improving the expression ability of the feature map;secondly,the weighted sum of CrossEntropy Loss and Dice loss is designed as the mask loss to improve the segmentation accuracy and robustness of the model;finally,the non-maximal suppression(NMS)algorithm to better handle the overlap problem in instance *** results show that the proposed method achieves an average segmentation accuracy of mAP of 80.4% on three types of electrical equipment datasets,including transformers,insulators and voltage transformers,which improve the detection accuracy by more than 5.7% compared with the original Solov2 *** segmentation model proposed can provide a focusing technical means for the intelligent management of power systems.

关键词： Deep learning electrical equipment attention mechanism weighted loss function

A Layered Energy-Efficient Multi-Node Scheduling Mechanism for Large-Scale WSN

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第4期79卷 1335-1351页

作者： Xue Zhao Shaojun tao hongying Tang jiang Wang Baoqing Li Science and Technology on Microsystem Laboratory Shanghai Institute of Microsystem and Information TechnologyChinese Academy of SciencesShanghai200050China

In recent years, target tracking has been considered one of the most important applications of wireless sensornetwork (WSN). Optimizing target tracking performance and prolonging network lifetime are two equally criticalobjectives in this scenario. The existing mechanisms still have weaknesses in balancing the two demands. Theproposed heuristic multi-node collaborative scheduling mechanism (HMNCS) comprises cluster head (CH)election, pre-selection, and task set selectionmechanisms, where the latter two kinds of selections forma two-layerselection mechanism. The CH election innovatively introduces the movement trend of the target and establishesa scoring mechanism to determine the optimal CH, which can delay the CH rotation and thus reduce energyconsumption. The pre-selection mechanism adaptively filters out suitable nodes as the candidate task set to applyfor tracking tasks, which can reduce the application consumption and the overhead of the following task setselection. Finally, the task node selection is mathematically transformed into an optimization problem and thegenetic algorithm is adopted to form a final task set in the task set selection mechanism. Simulation results showthat HMNCS outperforms other compared mechanisms in the tracking accuracy and the network lifetime.

关键词： Node scheduling pre-selection target tracking WSN

Addressing chemoresistance with a lipid gemcitabine nanotherapeutic strategy for effective treatment of pancreatic cancer

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Nano Research 2024年第9期17卷 8377-8388页

作者： Jiawei hong Shiyun Xian Shusen Zheng Hangxiang Wang Donghai jiang The First Affiliated Hospital NHC Key Laboratory of Combined Multi-Organ TransplantationCollaborative Innovation Center for Diagnosis and Treatment of Infectious DiseasesState Key Laboratory for Diagnosis and Treatment of Infectious DiseasesSchool of MedicineZhejiang UniversityHangzhou 310003China Jinan Microecological Biomedicine Shandong Laboratory Jinan 250117China

Resistance to gemcitabine in pancreatic cancer poses a significant clinical *** investigation is warranted to assess whether nano-formulation strategy can be employed to enhance the sensitivity of resistant strains to gemcitabine *** this study,using gemcitabine-resistant pancreatic cancer cell lines,we examined the therapeutic potential of a gemcitabine nanodelivery platform and assessed the ability to overcome drug resistance against resistant *** of human equilibrative nucleoside transporter 1(hENT1)led to reduced cellular uptake of gemcitabine,resulting in chemoresistance in pancreatic *** nanoparticles circumvented the entry blockade caused by hENT1 silencing through *** entry via clathrin-mediated endocytosis increased intracellular gemcitabine accumulation in gemcitabine-resistant pancreatic cancer ***,gemcitabine nanoparticles are preferential in vivo delivery to tumor tissues,likely due to the enhanced permeability and retention *** comparison to free gemcitabine,gemcitabine nanoparticles demonstrate a more pronounced cytotoxic effect on gemcitabine-resistant pancreatic cancer cells,with favorable *** study improved the efficacy of gemcitabine through nanotechnology,providing a novel strategy to address gemcitabine-resistant pancreatic cancer.

关键词： pancreatic cancer gemcitabine resistance gemcitabine nanoparticle human equilibrative nucleoside transporter 1

Quantitative effect of kerogen type on the hydrocarbon generation potential of Paleogene lacustrine source rocks,Liaohe Western Depression,China

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Petroleum Science 2024年第1期21卷 14-30页

作者： Sha-Sha Hui Xiong-Qi Pang Fu-Jie jiang Chen-Xi Wang Shu-Xing Mei tao Hu hong Pang Min Li Xiao-Long Zhou Kan-Yuan Shi State Key Laboratory of Petroleum Resources and Prospecting China University of Petroleum(Beijing)Beijing102249China College of Geosciences China University of Petroleum(Beijing)Beijing102249China Exploration and Development Research Institute PetroChina Liaohe Oilfield CompanyPanjin 124010LiaoningChina

Kerogen types exert a decisive effect on the onset and capacity of hydrocarbon generation of source *** source rocks in the Liaohe Western Depression are characterized by thick deposition,high total organic carbon(TOC)content,various kerogen types,and a wide range of thermal ***,their hydrocarbon generation potential and resource estimation can be *** this study,geochemical tests,numerical analysis,hydrocarbon generation kinetics,and basin modeling were integrated to investigate the differential effects of kerogen types on the hydrocarbon generation potential of lacustrine source *** hydrocarbon generation and expulsion(HGE)models of different kerogen types were established quantitatively upon abundant Rock-Eval/TOC/vitrinite reflectance(R_(o))*** sets of good-excellent source rocks deposited in the fourth(Es4),third(Es3),and first(Es1)members of Paleogene Shahejie Formation,are predominantly types I-II_(1),II_(1)-II_(2),and II-III,*** activation energy of types I-II_(2)kerogen is concentrated(180-230 kcal/mol),whereas that of type III kerogen is widely distributed(150-280 kcal/mol).The original hydrocarbon generation potentials of types I,II_(1),II_(2),and III kerogens are 790,510,270,and 85 mg/g TOC,*** Ro values of the hydrocarbon generation threshold for type I-III source rocks gradually increase from 0.42%to 0.74%,and Ro values of the hydrocarbon expulsion threshold increase from 0.49%to 0.87%.Types I and II_(1)source rocks are characterized by earlier hydrocarbon generation,more rapid hydrocarbon expulsion,and narrower hydrocarbon generation windows than types II_(2)and III source *** kerogen types also affect the HGE history and resource *** types(conventional,tight,and shale oil/gas)and three levels(realistic,expected,and prospective)of hydrocarbon resources of different members in the Liaohe Western Depression are *** suggest that t

关键词： Kerogen type Hydrocarbon generation potential Lacustrine source rocks Liaohe western depression

维普期刊数据库

High-precision nondestructive evaluation of a thermal barrier coating based on perovskite quantum dot anion exchange

在线全文

学校读者我要写书评

暂无评论

Nano Research 2024年第5期17卷 4582-4592页

作者： tao Han Shufang Ding Zifan Wang Sirong jiang Pengjiang Jing Tianshang Yi Yaqi Chen Chunzhi jiang Xiaofeng Zhang Microelectronics and Optoelectronics Technology Key Laboratory of Hunan Higher Education School of Physics and Electronic Electrical EngineeringXiangnan UniversityChenzhou 423000China National Engineering Laboratory for Modern Materials Surface Engineering Technology&The Key Lab of Guangdong for Modern Surface Engineering Technology Guangdong Institute of New MaterialsGuangdong Academy of SciencesGuangzhou 510650China

Thermal barrier coatings(TBCs)in gas turbine engines are used in expressly harsh environments;thus,assessing TBC integrity status is critical for safety and ***,traditional periodic maintenance involves visual inspections of the TBCs,requiring the gas turbine to be decommissioned and partially *** importantly,tiny defects or internal damages that easily cause coating failure cannot be *** this work,a new nondestructive evaluation(NDE)technique of TBCs based on quantum dot(QD)anion exchange is first explored *** exchanging anions between the Cl ions and the CsPbBr_(3) QDs,the degrees of salt corrosion of the TBCs are *** resultant NDE technique shows that the colour of the TBCs obviously changes from green to blue,accompanied by a large blueshift(~100 nm)of the photoluminescence(PL)peak *** addition,the relationship between the PL peak position and coating thermophysical properties indicates that the precision of this NDE technique may easily identify theμm-level of the thermal growth oxide(TGO)changes inside the TBCs.

关键词： thermal barrier coatings nondestructive evaluation CsPbBr_(3)quantum dot anion exchange thermal growth oxide

Channel modeling for MI-based wireless underground sensor networks with conductive objects

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7期67卷 333-334页

作者： Shuhan DENG Guanghua LIU Ziwei CHEN Huaijin ZHANG tao jiang Research Center of 6G Mobile Communications School of Cyber Science and EngineeringHuazhong University of Science and Technology School of Electronic Information and Communications Huazhong University of Science and Technology

Wireless underground sensor networks (WUSNs) are an emerging research field with promising applications. However, WUSNs links suffer from significant attenuation as the propagation of the signal needs to pass through soil, rock,and water. Magnetic induction (MI)-based transmission is proposed to overcome these issues as MI signal can penetrate most substances similarly without experiencing high path loss, which ensures the robustness of WUSNs links [1].

关键词： Wireless sensor networks

Global-local combined features to detect pain intensity from facial expression images with attention mechanism

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Electronic Science and Technology 2024年第3期22卷 80-93页

作者： jiang Wu Yi Shi Shun Yan hong-Mei Yan Yangtze Delta Region Institute(Huzhou) University of Electronic Science and Technology of ChinaHuzhou313001China MOE Key Laboratory for Neuroinformation School of Life Science and TechnologyUniversity of Electronic Science and Technology of ChinaChengdu611731China College of Engineering University of California at Santa BarbaraSanta Barbara93106USA

The estimation of pain intensity is critical for medical diagnosis and treatment of *** the development of image monitoring technology and artificial intelligence,automatic pain assessment based on facial expression and behavioral analysis shows a potential value in clinical *** paper reports a framework of convolutional neural network with global and local attention mechanism(GLA-CNN)for the effective detection of pain intensity at four-level thresholds using facial expression ***-CNN includes two modules,namely global attention network(GANet)and local attention network(LANet).LANet is responsible for extracting representative local patch features of faces,while GANet extracts whole facial features to compensate for the ignored correlative features between *** the end,the global correlational and local subtle features are fused for the final estimation of pain *** under the UNBC-McMaster Shoulder Pain database demonstrate that GLA-CNN outperforms other state-of-the-art ***,a visualization analysis is conducted to present the feature map of GLA-CNN,intuitively showing that it can extract not only local pain features but also global correlative facial *** study demonstrates that pain assessment based on facial expression is a non-invasive and feasible method,and can be employed as an auxiliary pain assessment tool in clinical practice.

关键词： Attention Convolutional neural network Facial expression Pain intensity