检索结果-南通市图书馆

Learnware:small models do big

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第1期67卷 39-50页

作者： Zhi-hua zhou Zhi-Hao TAN National Key Laboratory for Novel Software Technology Nanjing University

There are complaints about current machine learning techniques such as the requirement of a huge amount of training data and proficient training skills,the difficulty of continual learning,the risk of catastrophic forgetting,and the leaking of data privacy/*** research efforts have been focusing on one of those concerned issues separately,paying less attention to the fact that most issues are entangled in *** prevailing big model paradigm,which has achieved impressive results in natural language processing and computer vision applications,has not yet addressed those issues,whereas becoming a serious source of carbon *** article offers an overview of the learnware paradigm,which attempts to enable users not to need to build machine learning models from scratch,with the hope of reusing small models to do things even beyond their original purposes,where the key ingredient is the specification which enables a trained model to be adequately identified to reuse according to the requirement of future users who know nothing about the model in advance.

关键词： artificial intelligence machine learning learnware

Predicting the International Roughness Index of JPCP and CRCP Rigid Pavement:A Random Forest(RF)Model Hybridized with Modified Beetle Antennae Search(MBAS)for Higher Accuracy

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & Sciences 2024年第5期139卷 1557-1582页

作者： zhou ji Mengmeng zhou Qiang Wang jiandong huang College of Civil and Environmental Engineering Hunan University of Science and EngineeringYongzhou425006China School of Mines China University of Mining and TechnologyXuzhou221116China School of Civil Engineering Guangzhou UniversityGuangzhou510006China

To improve the prediction accuracy of the International Roughness Index(IRI)of Jointed PlainConcrete Pavements(JPCP)and Continuously Reinforced Concrete Pavements(CRCP),a machine learning approach is developed in this study for the modelling,combining an improved Beetle Antennae Search(MBAS)algorithm and Random Forest(RF)*** 10-fold cross-validation was applied to verify the reliability and accuracy of the model proposed in this *** importance scores of all input variables on the IRI of JPCP and CRCP were analysed as *** results by the comparative analysis showed the prediction accuracy of the IRI of the newly developed MBAS and RF hybrid machine learning model(RF-MBAS)in this study is higher,indicated by the RMSE and R values of 0.2732 and 0.9476 for the JPCP as well as the RMSE and R values of 0.1863 and 0.9182 for the *** accuracy of this obtained result far exceeds that of the IRI prediction model used in the traditional Mechanistic-Empirical Pavement Design Guide(MEPDG),indicating the great potential of this developed *** importance analysis showed that the IRI of JPCP and CRCP was proportional to the corresponding input variables in this study,including the total joint faulting cumulated per KM(TFAULT),percent subgrade material passing the 0.075-mm Sieve(P_(200))and pavement surface area with flexible and rigid patching(all Severities)(PATCH)which scored higher.

关键词： Cement pavement JPCP CRCP RF-MBAS IRI

Enhanced thermoelectric performance enabled by compositing ZrO_(2)in n-type SiGe alloy with low thermal conductivity

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Rare Metals 2024年第3期43卷 1167-1176页

作者： Meng-Fei Wang hua-Jun Lai ji-Sheng Liang Jun-Liang Chen Wang-Yang Ding Qi zhou Ying Peng Cheng-Yan Liu Lei Miao Guangxi Key Laboratory for Relativity Astrophysics State Key Laboratory of Featured Metal Materials and Life-Cycle Safety for Composite StructuresSchool of Physical Science and TechnologyGuangxi UniversityNanning530004China Guangxi Academy of Sciences Nanning530007China Key Laboratory of Information Material Engineering Research Center of Electronic Information Materials and DevicesGuilin University of Electronic TechnologyMinistry of EducationGuilin541004China Guangxi Key Laboratory of Precision Navigation Technology and Application School of Information and CommunicationGuilin University of Electronic TechnologyGuilin541004China

SiGe-based thermoelectric(TE)materials have gained increasing interests due to their low maintenance costs,environmental friendliness and long ***,the intrinsically high thermal conductivity of Si-based materials also results in poor TE *** this investigation,a zirconia(ZrO_(2))composite strategy was applied to an n-type SiGe alloy,tremendously elevating its TE *** mechanical alloying and spark plasma sintering(SPS)processes,the ZrO_(2)induced the formation of nanopores in the SiGe matrix via phosphorus ***,such increase in porosity enhanced the phonon scattering and dramatically suppressed lattice thermal conductivity,from 2.83 to 1.59 W·m^(-1)·K^(-1)at 873 ***,reduced phosphorus doping led to an increase in Seebeck coefficients and a relatively minor decrease in electrical conductivity,The power factor didn't deteriorate significantly,either,as its maximum of~3.43 mW·m^(-1-)K^(-2)was achieved at 873 K with(Si_(0.8)Ge_(0.2))_(0.097)P_(0.03)(ZrO_(2))_(0.003).In short,a peak figure of merit(ZT)of~1.27 at 873 K and an average ZT~0.7 from 323 to 873 K were *** study demonstrates that the electrical and thermal transportation of SiGe material can be synergistically tuned by compositing ZrO_(2),illustrating a novel strategy to optimize the TE properties of bulk materials.

关键词： Thermoelectric SiGe alloy ZrO2 composting P element adsorption nanopores

Quasi-shadowing for Z^(d)-actions

在线全文

学校读者我要写书评

暂无评论

Acta Mathematica Sinica,English Series 2024年第6期40卷 1563-1580页

作者： Juan PAN Xian Kun REN Yun hua zhou School of Mathematics and Statistics Chongqing Technology and Business UniversityChongqing 400067P.R.China College of Mathematics and Statistics Chongqing UniversityChongqing 401331P.R.China

A diffeomorphism is non-uniformly partially hyperbolic if it preserves an ergodic measure with at least one zero Lyapunov *** prove that a C^(1)-smooth Z^(d)-action has the quasishadowing property if one of the genera... 详细信息

关键词： Quasi-shadowing Z^(d)-action non-uniformly partially hyperbolic ergodic measure

Mechanism of local solidification time variations with melt rate during vacuum arc remelting process of 8Cr4Mo4V high-strength steel

在线全文

学校读者我要写书评

暂无评论

Journal of Iron and Steel Research(International) 2024年第2期31卷 377-388页

作者： Tao Pan Hong-chun Zhu zhou-hua jiang hua-bing Li Rui Zhang Hao Feng School of Metallurgy Northeastern UniversityShenyang110819LiaoningChina State Key Laboratory of Rolling and Automation Northeastern UniversityShenyang110819LiaoningChina

A 2D axisymmetric numerical model was established to investigate the variations of molten pool with different melt rates during the vacuum arc remelting of 8Cr4Mo4V high-strength steel,and the ingot growth was simulated by dynamic mesh *** results show that as the ingot grows,the molten pool profile changes from shallow and flat to V-shaped,and both the molten pool depth and the mushy width ***,the variation of both the molten pool shape and the mushy width melt rate is clarified by the thermal equilibrium *** melt rate increases,both the molten pool depth and the mushy width *** is caused by the increment in sensible heat stored in the ingot due to the limitation of the cooling capacity of the *** nonlinear increment in sensible heat leads to a nonlinear increase in the mushy *** addition,as melt rate increases,the local solidification time(LST)of ingot decreases obviously at first and then *** melt rate is controlled in a suitable range,LST is the lowest and the secondary dendrite arm spacing of the ingot is the smallest,which can effectively improve the compactness degree of 8Cr4Mo4V high-strength steel.

关键词： 8Cr4Mo4V high-strength steel Vacuum arc remelting process Melt rate Molten pool Local solidification time

Structure and 2p decay mechanism of^(18)Mg

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Nuclear Science and Techniques 2024年第6期35卷 145-154页

作者： Long zhou De-Qing Fang Si-Min Wang Hui hua Key Laboratory of Nuclear Physics and Ion-beam Application(MOE) Institute of Modern PhysicsFudan UniversityShanghai 200433China Shanghai Research Center for Theoretical Nuclear Physics NSFC and Fudan UniversityShanghai 200438China School of Physics and State Key Laboratory of Nuclear Physics and Technology Peking UniversityBeijing 100871China

The recently discovered,extremely proton-rich nuclide 18 Mg exhibits ground-state decay via two sequential two-proton(2p)emissions through the intermediate nucleus,*** study investigates the structure and the initial 2p decay mechanism of^(18)Mgby examining the density and correlations of the valence protons using a three-body Gamow coupled-channel *** results show that the ground state of^(18)Mgis significantly influenced by the continuum,resulting in a significant s-wave ***,based on the current framework,this does not lead to a significant deviation in mirror symmetry in either the structure or spectroscopy of the 18Mg-18C ***,the time evolution analysis of the^(18)Mgground state suggests a simultaneous 2p emission during the first step of *** observed nucleon-nucleon correlations align with those of the light-mass 2p emitters,indicating a consistent decay behavior within this nuclear region.

关键词： Structure of^(18)Mg Two-proton decay Proton-proton correlation

A reweighted damped singular spectrum analysis method for robust seismic noise suppression

在线全文

学校读者我要写书评

暂无评论

Petroleum Science 2024年第3期21卷 1671-1682页

作者： Wei-Lin huang Yan-Xin zhou Yang zhou Wei-jie Liu ji-Dong Li State Key Laboratory of Petroleum Resources and Engineering China University of Petroleum(Beijing)Beijing2249China SINOPEC Petroleum Exploration and Production Research Institute Beijing102206China SINOPEC Geophysical Research Institute Nanjing211103jiangsuChina

(Multichannel)Singular spectrum analysis is considered as one of the most effective methods for seismic incoherent noise *** utilizes the low-rank feature of seismic signal and regards the noise suppression as a low-rank reconstruction ***,in some cases the seismic geophones receive some erratic disturbances and the amplitudes are dramatically larger than other *** presence of this kind of noise,called erratic noise,makes singular spectrum analysis(SSA)reconstruction unstable and has undesirable effects on the final *** robustify the low-rank reconstruction of seismic data by a reweighted damped SSA(RD-SSA)*** incorporates the damped SSA,an improved version of SSA,into a reweighted *** damping operator is used to weaken the artificial disturbance introduced by the low-rank projection of both erratic and random *** central idea of the RD-SSA method is to iteratively approximate the observed data with the quadratic norm for the first iteration and the Tukeys bisquare norm for the rest *** RD-SSA method can suppress seismic incoherent noise and keep the reconstruction process robust to the erratic *** feasibility of RD-SSA is validated via both synthetic and field data examples.

关键词： Singular spectrum analysis Damping operator Seismic erratic noise Seismic signal processing Robust low-rank reconstruction

Evaluation of the value of combined detection of tumor markers CA724,carcinoembryonic antigen,CA242,and CA19-9 in gastric cancer

在线全文

学校读者我要写书评

暂无评论

World Journal of Gastrointestinal Oncology 2024年第5期16卷 1737-1744页

作者： Chong-Mei zhou Shao-hua Zhao Department of Clinical Laboratory Henan Provincial People's HospitalZhengzhou 450003Henan ProvinceChina

BACKGROUND Gastric cancer is a global health concern that poses a significant threat to human *** To detecting serum changes in carcinoembryonic antigen(CEA),carbohydrate antigens(CA)724,CA242,and CA19-9 expression among patients with gastric *** Eighty patients diagnosed with gastric cancer between January 2020 and January 2023 were included in the observation group,while 80 patients with benign gastric diseases were included in the control *** groups were tested for tumor markers(CA724,CEA,CA242,and CA19-9].Tumor marker indicators(CA724,CEA,CA242,and CA19-9)were compared between the two groups,assessing positive rates of tumor markers across various stages in the observation ***,single and combined detection of various tumor markers were *** The sensitivity,specificity,accuracy,positive predictive value,and negative predictive value observed for the combined detection of CA724,CEA,CA242,and CA19-9 were higher than those of CA724,CEA,CA242,and CA19-9 ***,the combined detection of CA724,CEA,CA242,and CA19-9 has a high diagnostic accuracy and could reduce the occurrence of missed or misdiagnosed cases,facilitating the early diagnosis and treatment of *** CA724,CEA,CA242,and CA19-9 serum levels in gastric cancer patients significantly surpassed those in non-gastric cancer patients(P<0.05).Their combined detection can improve the diagnostic accuracy for gastric cancer,warranting clinical promotion.

关键词： Biomarkers Serum Gastric cancer Diagnostic value

Underwater Image Classification Based on EfficientnetB0 and Two-Hidden-Layer Random Vector Functional Link

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Ocean University of China 2024年第2期23卷 392-404页

作者： zhou Zhiyu LIU Mingxuan ji Haodong WANG Yaming ZHU Zefei School of Information Science and Technology Zhejiang Sci-Tech UniversityHangzhou 310018China Zhejiang Key Laboratory of DDIMCCP Lishui UniversityLishui 323000China School of Mechanical Engineering Hangzhou Dianzi UniversityHangzhou 310018China

The ocean plays an important role in maintaining the equilibrium of Earth’s ecology and providing humans access to a wealth of *** obtain a high-precision underwater image classification model,we propose a classification model that combines an EfficientnetB0 neural network and a two-hidden-layer random vector functional link network(EfficientnetB0-TRVFL).The features of underwater images were extracted using the EfficientnetB0 neural network pretrained via ImageNet,and a new fully connected layer was trained on the underwater image dataset using the transfer learning *** learning ensures the initial performance of the network and helps in the development of a high-precision classification ***,a TRVFL was proposed to improve the classification property of the *** construction of the two hidden layers exhibited a high accuracy when the same hidden layer nodes were *** parameters of the second hidden layer were obtained using a novel calculation method,which reduced the outcome error to improve the performance instability caused by the random generation of parameters of ***,the TRVFL classifier was used to classify features and obtain classification *** proposed EfficientnetB0-TRVFL classification model achieved 87.28%,74.06%,and 99.59%accuracy on the MLC2008,MLC2009,and Fish-gres datasets,*** best convolutional neural networks and existing methods were stacked up through box plots and Kolmogorov-Smirnov tests,*** increases imply improved systematization properties in underwater image classification *** image classification model offers important performance advantages and better stability compared with existing methods.

关键词： underwater image classification EfficientnetB0 random vector functional link convolutional neural network