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

Probabilistic seismic inversion based on physics-guided deep mixture density network

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Petroleum Science 2024年第3期21卷 1611-1631页

作者： Qian-hao Sun Zhao-yun Zong Xin Li National Key Laboratory of Deep Oil and Gas China University of Petroleum(East China)Qingdao266580ShandongChina Laoshan Laboratory Qingdao266580ShandongChina CNOOC Research Institute Ltd. Beijing100028China

Deterministic inversion based on deep learning has been widely utilized in model parameters *** by logging data,seismic data,wavelet and modeling operator,deterministic inversion based on deep learning can establish nonlinear relationships between seismic data and model ***,seismic data lacks low-frequency and contains noise,which increases the non-uniqueness of the *** conventional inversion method based on deep learning can only establish the deterministic relationship between seismic data and parameters,and cannot quantify the uncertainty of *** order to quickly quantify the uncertainty,a physics-guided deep mixture density network(PG-DMDN)is established by combining the mixture density network(MDN)with the deep neural network(DNN).Compared with Bayesian neural network(BNN)and network dropout,PG-DMDN has lower computing cost and shorter training time.A low-frequency model is introduced in the training process of the network to help the network learn the nonlinear relationship between narrowband seismic data and low-frequency *** addition,the block constraints are added to the PG-DMDN framework to improve the horizontal continuity of the inversion *** illustrate the benefits of proposed method,the PG-DMDN is compared with existing semi-supervised inversion *** synthetic data examples of Marmousi II model are utilized to quantify the influence of forward modeling part,low-frequency model,noise and the pseudo-wells number on inversion results,and prove the feasibility and stability of the proposed *** addition,the robustness and generality of the proposed method are verified by the field seismic data.

关键词： Deep learning Probabilistic inversion Physics-guided Deep mixture density network

Effect of the vortical structures on the hydrodynamic performance of a pitching hydrofoil

在线全文

学校读者我要写书评

暂无评论

Journal of Hydrodynamics 2024年第2期36卷 406-420页

作者： Rui Yuan Hui-yun hao Qin Wu yun-qing Liu Biao Huang School of Mechanical Engineering Beijing Institute of TechnologyBeijing100081China

The objective is to study the vortical structural behaviors of a transient pitching hydrofoil and their effects on the hydrodynamic performance. The pitching motion of the hydrofoil is set to pitch up with an almost constant rate from 5° to 15° and then back to 5°, with the Reynolds number 4.4×10^(5) and the frequency 2 Hz. The results show that the main coherent structures around the pitching hydrofoil include small-scale laminar separation bubble (LSB), large-scale second vortex (SV) and trailing edge vortex (TEV) which are all vortical. The relationship between the vortical structure and the lift is investigated with the finite-domain impulse theory. It indicates that the major part of the lift is contributed by the LSB, whereas the shedding and the formation of the SV and TEV cause the fluctuation of the lift. The proper orthogonal decomposition (POD) method is applied to capture the most energetic modes, revealing that the LSB mode occupies a large amount of energy in the flow field. The dynamic mode decomposition (DMD) method accurately extracts the dominant frequency and modal characteristics, with the first mode corresponding to the mean flow, the second mode corresponding to the LSB structure and the third and fourth modes corresponding to the vortex shedding.

关键词： Hydrodynamics vortical structures finite-domain impulse theory proper orthogonal decomposition(POD) dynamic mode decomposition(DMD) Pitching hydrofoil

New material of Coniopteris simplex from the Middle Jurassic of the Ordos Basin,Inner Mongolia,China and implications on its spatio-temporal distribution and paleogeography

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 Palaeogeography 2024年第2期13卷 199-211页

作者： yun-Feng Li Chang-Lu Zhang Fei Liang Xiao tan Fan-hao Gong Chun-Lin Sun Tao Li Yu-Ling Na Research Center of Paleontology and Stratigraphy Jilin UniversityChangchun 130026Jilin ProvinceChina College of Earth Sciences Jilin UniversityChangchun 130061Jilin ProvinceChina College of Paleontology Shenyang Normal UniversityShenyang 110034Liaoning ProvinceChina Key Lab of Evolution of Past Life in NE Asia Ministry of Natural ResourcesShenyang Normal UniversityShenyang 110034Liaoning ProvinceChina Shenyang Center of Geological Survey China Geological SurveyShenyang 110034Liaoning ProvinceChina

Coniopteris simplex is a common component in many Jurassic ***,due to morphological variations in sterile fronds and incomplete preservation of reproductive organs,its classification has been controversial for a long ***,we present new material collected from the Middle Jurassic Yan’an Formation in the Ordos Basin,Inner *** new material reveals important morphological characters,including hemi-dimorphic fronds,two forms of sterile pinnae,short-stalked sori and sporangia on fertile fronds,annulus of sporangium composed of about 30 cells,in situ rounded triangular spores with protruded suture,and the lips of trilete laesura encircled by one or two rows of small *** on these observations,we emended the diagnosis of this common *** spatio-temporal distribution of Coniopteris simplex shows that the iconic species was restricted to the paleolatitudes ranging from 19.5°N to 48.9°N in the Northern Hemisphere during the Early—Late Jurassic,mainly distributed in the tropical to paratropical zones,and the paleolongitude ranging from 9.6°E to 129.3°E,which may be limited by the opening of the Viking Corridor and the splitting of the Central Atlantic Ocean.

关键词： Coniopteris simplex Yan'an Formation Middle Jurassic Inner Mongolia Paleogeography

Fabrication of corrosion inhibiting microcapsules functional coating and their high-efficiency corrosion resistance

在线全文

学校读者我要写书评

暂无评论

Rare Metals 2024年第2期43卷 842-848页

作者： hao-Ran yun Lu Lu Xue-Long hao Xing-Xiang Zhang School of Material Science and Engineering Tiangong UniversityTianjin300387China National Center of Analysis and Testing for Nonferrous Metals and Electronic Materials General Research Institute for Nonferrous MetalsBeijing100088China Guobiao(Beijing)Testing&Certification Co. Ltd.Beijing101407China

Benzotriazole(BTA)was loaded into calcium carbonate microsphere shell by self-assembly method and a new type of high-efficiency corrosion inhibiting microcapsules(CIMs)were fabricated,which were added and dispersed into epoxy resin coating to obtain a new corrosion resistance functional *** corrosion inhibition effect of benzotriazole on aluminum alloy matrix was verified by polarization curve *** effectiveness of the CIMs and their loading capacity on inhibitor were evaluated by scanning electron microscope,laser confocal Raman spectroscopy together with thermogravimetric analysis.

关键词： corrosion coating resistance

Surface roughness classification using light scattering matrix and deep learning

在线全文

学校读者我要写书评

暂无评论

Science China(Technological Sciences) 2024年第2期67卷 520-535页

作者： SUN hao tan Wei RUAN YiXiao BAI Long XU JianFeng State Key Laboratory of Intelligent Manufacturing Equipment and Technology School of Mechanical Science and EngineeringHuazhong University of Science and TechnologyWuhan 430074China Wuhan Digital Design and Manufacturing Innovation Center Co. LtdWuhan 430074China

High-quality optical components have been widely used in various applications;thus,extremely high beam quality is ***,surface roughness is a key indicator of the surface *** this study,the angular distribution of light scattering field intensity was obtained for surfaces having different roughness profiles based on the finite difference time domain(FDTD)method,and the results were compared with those obtained using the generalized Harvey-Shack(GHS)*** was shown that the FDTD approach can be used for an accurate simulation of the scattered field of a rough surface,and the superposition of results obtained from many surfaces that have the same roughness level was in good agreement with the result given by the analytic GHS model.A light scattering matrix(LSM)method was proposed based on the FDTD simulation results that could obtain rich surface roughness *** classification effect of LSM was compared with that of the single-incidence scattering distribution(SISD)based on a ResNet-50 deep learning *** classification accuracy of the model trained with the LSM dataset was obtained as 95.74%,which was 23.40%higher than that trained using the SISD ***,the effects of different noise types and filtering methods on the classification performance were analyzed,and the LSM was also shown to improve the robustness and generalizability of the trained surface roughness ***,the proposed LSM method has important implications for improving the data acquisition scheme of current light scattering measurement systems,and it also has the potential to be used for detection and characterization of surface defects of optical components.

关键词： surface roughness FDTD simulation GHS theory deep learning light scattering matrix

Emerging clinical relevance of microbiome in cancer:promising biomarkers and therapeutic targets

在线全文

学校读者我要写书评

暂无评论

Protein & Cell 2024年第4期15卷 239-260页

作者： Jia-hao Dai Xi-Rong tan Han Qiao Na Liu State Key Laboratory of Oncology in South China Guangdong Key Laboratory of Nasopharyngeal Carcinoma Diagnosis and TherapyGuangdong Provincial Clinical Research Center for CancerSun Yat-sen University Cancer CenterGuangzhou 510050China

The profound influence of microbiota in cancer initiation and progression has been under the spotlight for years,leading to numerous researches on cancer microbiome entering clinical *** promising biomarkers and therapeutic targets,the critical involvement of microbiota in cancer clinical practice has been increasingly ***,recent progress in this field is *** describe the potential of tumor-associated microbiota as effective diagnostic and prognostic biomarkers,*** addition,we highlight the relationship between microbiota and the therapeutic efficacy,toxicity,or side effects of commonly utilized treatments for cancer,including chemotherapy,radiotherapy,and *** that microbial factors influence the cancer treatment outcome,we further summarize some dominating microbial interventions and discuss the hidden risks of these *** review aims to provide an overview of the applications and advancements of microbes in cancer clinical relevance.

关键词： microbiome cancer diagnosis prognosis therapy

Considerations of advanced aqueous batteries recycling:A perspective

在线全文

学校读者我要写书评

暂无评论

Journal of Energy Chemistry 2024年第6期93卷 46-54,I0002页

作者： yun Zhao Yuqiong Kang hao Du Feiyu Kang Baohua Li Institute of Materials Research Tsinghua Shenzhen International Graduate SchoolTsinghua UniversityShenzhen 518055GuangdongChina

*** Battery technology,serving as a basis for a rechargeable society,has developed rapidly over a century[1].The progress can be roughly divided into three stages,with the initial two mainly focusing on aqueous battery *** to 1859,research primarily focused on the conceptualization of batteries and the development of primary(non-rechargeable)cells[2].In the subsequent150 years,rechargeable battery technology emerged。

关键词： battery Battery primarily

Rethinking multi-spatial information for transferable adversarial attacks on speaker recognition systems

在线全文

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence Technology 2024年第3期9卷 620-631页

作者： Junjian Zhang hao tan Le Wang Yaguan Qian Zhaoquan Gu Guangzhou University GuangzhouChina Department of New Networks Peng Cheng LaboratoryShenzhenChina Harbin Institute of Technology(Shenzhen) ShenzhenChina Zhejiang University of Science and Technology HangzhouChina

Adversarial attacks have been posing significant security concerns to intelligent systems,such as speaker recognition systems(SRSs).Most attacks assume the neural networks in the systems are known beforehand,while black-box attacks are proposed without such information to meet practical *** black-box attacks improve trans-ferability by integrating multiple models or training on multiple datasets,but these methods are *** by the optimisation strategy with spatial information on the perturbed paths and samples,we propose a Dual Spatial Momentum Iterative Fast Gradient Sign Method(DS-MI-FGSM)to improve the transferability of black-box at-tacks against ***,DS-MI-FGSM only needs a single data and one model as the input;by extending to the data and model neighbouring spaces,it generates adver-sarial examples against the integrating *** reduce the risk of overfitting,DS-MI-FGSM also introduces gradient masking to improve *** authors conduct extensive experiments regarding the speaker recognition task,and the results demonstrate the effectiveness of their method,which can achieve up to 92%attack success rate on the victim model in black-box scenarios with only one known model.

关键词： speaker recognition spoofing attacks