检索结果-南通市图书馆

FSCIL-eACA: Few-Shot Class-Incremental Learning Network Based on embedding Augmentation and Classifier Adaptation for Image Classification

引用

Chinese Journal of electronics 2024年第1期33卷 139-152页

作者： Ruru ZHANG Haihong e Meina SONG School of Computer Science Beijing University of Posts and Telecommunications education Department Information Network engineering Research Center Beijing University of Posts and Telecommunications

The ability to learn incrementally is critical to the long-term operation of AI systems. Benefiting from the power of few-shot class-incremental learning(FSCIL), deep learning models can continuously recognize new classes with only a few samples. The difficulty is that limited instances of new classes will lead to overfitting and exacerbate the catastrophic forgetting of the old classes. Most previous works alleviate the above problems by imposing strong constraints on the model structure or parameters, but ignoring embedding network transferability and classifier adaptation(CA), failing to guarantee the efficient utilization of visual features and establishing relationships between old and new classes. In this paper, we propose a simple and novel approach from two perspectives: embedding bias and classifier bias. The method learns an embedding augmented(eA) network with cross-class transfer and class-specific discriminative abilities based on self-supervised learning and modulated attention to alleviate embedding bias. Based on the adaptive incremental classifier learning scheme to realize incremental learning capability,guiding the adaptive update of prototypes and feature embeddings to alleviate classifier bias. We conduct extensive experiments on two popular natural image datasets and two medical datasets. The experiments show that our method is significantly better than the baseline and achieves state-of-the-art results.

关键词： Training Adaptation models Visualization Image recognition Prototypes Self-supervised learning Power capacitors

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Dynamics of self-hybridized exciton–polaritons in 2D halide perovskites

引用

Light(Science & Applications) 2024年第1期13卷 13-22页

作者： Surendra B.Anantharaman Jason Lynch Christopher e.Stevens Christopher Munley Chentao Li Jin Hou Hao Zhang Andrew Torma thomas Darlington Francis Coen Kevin Li Arka Majumdar PJames Schuck Aditya Mohite Hayk Harutyunyan Joshua R.Hendrickson Deep Jariwala Department of electrical and Systems engineering University of PennsylvaniaPhiladelphiaPA19104USA KBR Inc. BeavercreekOH45431USA Air Force Research Laboratory Sensors DirectorateWright-Patterson Air Force BaseOH45433USA Department of Physics University of WashingtonSeattleWA98195USA Department of Physics Emory UniversityAtlantaGA30322USA Department of Chemical and Biomolecular engineering Rice UniversityHoustonTX77005USA Department of Materials Science and Nanoengineering Rice UniversityHoustonTX77005USA Applied Physics Program Smalley‐Curl InstituteRice UniversityHoustonTX77005USA Department of Mechanical engineering Columbia UniversityNew YorkNY10027USA Department of electrical and Computer engineering University of WashingtonSeattleWA98195USA

excitons,bound electron–hole pairs,in two-dimensional hybrid organic inorganic perovskites(2D HOIPs)are capable of forming hybrid light-matter states known as exciton-polaritons(e–Ps)when the excitonic medium is confined in an optical *** the case of 2D HOIPs,they can self-hybridize into e–Ps at specific thicknesses of the HOIP crystals that form a resonant optical cavity with the ***,the fundamental properties of these self-hybridized e–Ps in 2D HOIPs,including their role in ultrafast energy and/or charge transfer at interfaces,remain ***,we demonstrate that>0.5µm thick 2D HOIP crystals on Au substrates are capable of supporting multiple-orders of self-hybridized e–P *** e–Ps have high Q factors(>100)and modulate the optical dispersion for the crystal to enhance sub-gap absorption and *** varying excitation energy and ultrafast measurements,we also confirm energy transfer from higher energy e–Ps to lower energy e–***,we also demonstrate that e–Ps are capable of charge transport and transfer at *** findings provide new insights into charge and energy transfer in e–Ps opening new opportunities towards their manipulation for polaritonic devices.

关键词： charge polar excitation

来源：

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

The effect of the concentration of plastic waste on the formation of reaction products of the Ti–PeT system

引用

Green Chemical engineering 2024年第3期5卷 374-382页

作者： Aleksey e.Matveev Tomsk Scientific Center of the Siberian Branch of the Russian Academy of Sciences

The paper presents research results of the synthesis, phase composition, and structure of products obtained by highly exothermic reactions between Ti and C10H8O4mixture components, depending on the plastic waste concentration and Ti powder dispersiveness, density of initial samples, and synthesis medium. The dependence of the phase composition and structure of the synthesis products on the concentration of the polymer（plastic）component in the initial Ti-PeT(C10H8O4) system was found. When the plastic waste content is 20 wt% to 25 wt%, synthesis products contain TiC0.5O0.5–TiC0.6–0.75particle agglomerates. Further growth in the polyethylene terephthalate content from 30 wt% to 45 wt% leads to the formation of synthesis products consisting of titanium carbide(TiC0.6–1.0). The gaseous byproduct composition of the exothermic reaction is investigated for the Ti–-C10H8O4mixture composition. It is found that the gaseous by-product mostly contains hydrogen and carbon monoxide as well as impurities of different hydrocarbons（methane, acetylene, ethane, ethene） and carbon dioxide. The maximum adiabatic temperature of the gas combustion temperature is 2080℃, which demonstrates the possibility of using gas as a fuel for energy generation devices. Based on the data obtained, it is possible to create a basis for a new plastic waste technology to fabricate carbide-containing materials.

关键词： Plastic waste Titanium carbide Green technology Hydrogen synthesis

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

explainability and Interpretability in electric Load Forecasting Using Machine Learning Techniques–A Review

引用

energy and AI 2024年第2期16卷 483-496页

作者： Lukas Baur Konstantin Ditschuneit Maximilian Schambach Can Kaymakci thomas Wollmann Alexander Sauer Fraunhofer Institute for Manufacturing engineering and Automation IPA Nobelstraße 12Stuttgart70569Germany Institute for energy efficiency in Production(eeP) University of StuttgartNobelstraße 12Stuttgart70569Germany Merantix Momentum GmbH Max-Urich-Straße 3Berlin13355Germany

electric Load Forecasting(eLF)is the central instrument for planning and controlling demand response programs,electricity trading,and consumption *** to the increasing automation of these processes,meaningful and transparent forecasts become more and more ***,at the same time,the complexity of the used machine learning models and architectures *** there is an increasing interest in interpretable and explainable load forecasting methods,this work conducts a literature review to present already applied approaches regarding explainability and interpretability for load forecasts using Machine *** on extensive literature research covering eight publication portals,recurring modeling approaches,trends,and modeling techniques are identified and clustered by properties to achieve more interpretable and explainable load *** results on interpretability show an increase in the use of probabilistic models,methods for time series decomposition and the use of fuzzy logic in addition to classically interpretable *** explainable approaches are Feature Importance and Attention *** discussion shows that a lot of knowledge from the related field of time series forecasting still needs to be adapted to the problems in *** to other applications of explainable and interpretable methods such as clustering,there are currently relatively few research results,but with an increasing trend.

关键词： electric load forecasting explainability InterpretabilityStructured review

来源：

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

event-triggered H_(∞) consensus control for input-constrained multi-agent systems via reinforcement learning

引用

Control Theory and Technology 2024年第1期22卷 25-38页

作者： Jinxuan Zhang Chang-e Ren College of Information engineering Capital Normal UniversityBeijing100089China

This article presents an event-triggered H_(∞) consensus control scheme using reinforcement learning (RL) for nonlinear second-order multi-agent systems (MASs) with control constraints. First, considering control constraints, the constrained H_(∞) consensus problem is transformed into a multi-player zero-sum game with non-quadratic performance functions. Then, an event-triggered control method is presented to conserve communication resources and a new triggering condition is developed for each agent to make the triggering threshold independent of the disturbance attenuation level. To derive the optimal controller that can minimize the cost function in the case of worst disturbance, a constrained Hamilton–Jacobi–Bellman (HJB) equation is defined. Since it is difficult to solve analytically due to its strongly non-linearity, reinforcement learning (RL) is implemented to obtain the optimal controller. In specific, the optimal performance function and the worst-case disturbance are approximated by a time-triggered critic network;meanwhile, the optimal controller is approximated by event-triggered actor network. After that, Lyapunov analysis is utilized to prove the uniformly ultimately bounded (UUB) stability of the system and that the network weight errors are UUB. Finally, a simulation example is utilized to demonstrate the effectiveness of the control strategy provided.

关键词： H_(∞)optimal control Input constrains Multi-agent systems(MASs) Reinforcement learning(RL)

来源：

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

Self-Triggered Set Stabilization of Boolean Control Networks and Its Applications

引用

Ieee/CAA Journal of Automatica Sinica 2024年第7期11卷 1631-1642页

作者： Rong Zhao Jun-e Feng Dawei Zhang the School of Mathematics Shandong UniversityShandong UniversityJinan 250100China

Set stabilization is one of the essential problems in engineering systems, and self-triggered control(STC) can save the storage space for interactive information, and can be successfully applied in networked control systems with limited communication resources. In this study, the set stabilization problem and STC design of Boolean control networks are investigated via the semi-tensor product technique. On the one hand, the largest control invariant subset is calculated in terms of the strongly connected components of the state transition graph, by which a graph-theoretical condition for set stabilization is derived. On the other hand, a characteristic function is exploited to determine the triggering mechanism and feasible controls. Based on this, the minimum-time and minimum-triggering open-loop, state-feedback and output-feedback STCs for set stabilization are designed,respectively. As classic applications of self-triggered set stabilization, self-triggered synchronization, self-triggered output tracking and self-triggered output regulation are discussed as well. Additionally, several practical examples are given to illustrate the effectiveness of theoretical results.

关键词： Boolean control networks(BCNs) output regulation self-triggered control semi-tensor product of matrices set stabilization synchronization

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Nonlinear wave dispersion in monoatomic chains with lumped and distributed masses:discrete and continuum models

引用

Applied Mathematics and Mechanics(english edition) 2024年第4期45卷 633-648页

作者： e.GHAVANLOO S.eL-BORGI School of Mechanical engineering Shiraz UniversityShiraz 71963-16548Iran Mechanical engineering Program Texas A&M University at QatarDoha 23874Qatar

The main objective of this paper is to investigate the influence of inertia of nonlinear springs on the dispersion behavior of discrete monoatomic chains with lumped and distributed *** developed model can represent the wave propagation problem in a non-homogeneous material consisting of heavy inclusions embedded in a *** inclusions are idealized by lumped masses,and the matrix between adjacent inclusions is modeled by a nonlinear spring with distributed ***,the model is capable of depicting the wave propagation in bi-material bars,wherein the first material is represented by a rigid particle and the second one is represented by a nonlinear spring with distributed *** discrete model of the nonlinear monoatomic chain with lumped and distributed masses is first considered,and a closed-form expression of the dispersion relation is obtained by the second-order Lindstedt-Poincare method(LPM).Next,a continuum model for the nonlinear monoatomic chain is derived directly from its discrete lattice model by a suitable continualization *** subsequent use of the second-order method of multiple scales(MMS)facilitates the derivation of the corresponding nonlinear dispersion relation in a closed *** novelties of the present study consist of(i)considering the inertia of nonlinear springs on the dispersion behavior of the discrete mass-spring chains;(ii)developing the second-order LPM for the wave propagation in the discrete chains;and(iii)deriving a continuum model for the nonlinear monoatomic chains with lumped and distributed ***,a parametric study is conducted to examine the effects of the design parameters and the distributed spring mass on the nonlinear dispersion relations and phase velocities obtained from both the discrete and continuum *** parameters include the ratio of the spring mass to the lumped mass,the nonlinear stiffness coefficient of the spring,and the wave amplitude.

关键词： nonlinear mass-spring chain discrete model continuum model LindstedtPoincare method(LPM) method of multiple scales(MMS) dispersion phase velocity

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Annual variation of adult survival of a south-temperate House Wren population in Argentina

引用

Avian Research 2024年第2期15卷 229-237页

作者： Gustavo J.Fernandez Mariana e.Carro Paulo e.Llambías Grupo de Investigacion en ecofisiología de Fauna Silvestre INIBIOMA-CONICETAsentamiento Universitario San Martín de los Andes(AUSMA)Universidad Nacional Del ComahuePasaje de la Paz 235(8370)San Martín de los AndesNeuquenArgentina Grupo de Investigacion de Biología de Aves IADIZA CCT-MendozaUniversidad Nacional de CuyoAvenida RuizLeal s/n Parque General San Martin5500MendozaArgentina

Identifying factors affecting the survival of individuals is essential for understanding the evolution of life-history traits and population *** numerous studies on this subject in north-temperate environments,there is a lack of equivalent studies at similar latitudes in the ***,we used a 14-year dataset of capture,banding,and resighting to estimate the annual variation in the apparent adult survival probability of a south-temperate population of House Wrens(Troglodytes aedon bonariae).We evaluated temporal variation in sur-vival and the effect of environmental(climatic)and demographic variables(adult abundance,total number of fledglings produced during each breeding season)on survival *** found that the probability of adult survival decreased as the abundance of breeding adults *** density-dependent effect could be related to the resident lifestyle of southern House Wrens,which could determine an intense competition for territories and resources that ultimately would affect their survival.

关键词： Annual variation CJS Density-dependence Survival Troglodytes aedon bonariae

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Beyond p-y method:A review of artificial intelligence approaches for predicting lateral capacity of drilled shafts in clayey soils

引用

Journal of Rock Mechanics and Geotechnical engineering 2024年第9期16卷 3812-3840页

作者： M.e.Al-Atroush A.e.Aboelela ezz el-Din Hemdan Civil and environmental engineering Program Engineering Management DepartmentCollege of EngineeringPrince Sultan UniversityRiyadhSaudi Arabia Department of Structural engineering Faculty of EngineeringAin Shams UniversityCairoEgypt Structure and Materials Research Lab Prince Sultan UniversityRiyadhSaudi Arabia Computer Science and engineering Department Faculty of Electronic EngineeringMenoufia UniversityMenoufiaEgypt

In 2023,pivotal advancements in artificial intelligence(AI)have significantly *** that in mind,traditional methodologies,notably the p-y approach,have struggled to accurately model the complex,nonlinear soil-structure interactions of laterally loaded large-diameter drilled *** study undertakes a rigorous evaluation of machine learning(ML)and deep learning(DL)techniques,offering a comprehensive review of their application in addressing this geotechnical challenge.A thorough review and comparative analysis have been carried out to investigate various AI models such as artificial neural networks(ANNs),relevance vector machines(RVMs),and least squares support vector machines(LSSVMs).It was found that despite ML approaches outperforming classic methods in predicting the lateral behavior of piles,their‘black box'nature and reliance only on a data-driven approach made their results showcase statistical robustness rather than clear geotechnical insights,a fact underscored by the mathematical equations derived from these ***,the research identified a gap in the availability of drilled shaft datasets,limiting the extendibility of current findings to large-diameter *** extensive dataset,compiled from a series of lateral loading tests on free-head drilled shaft with varying properties and geometries,was introduced to bridge this *** paper concluded with a direction for future research,proposes the integration of physics-informed neural networks(PINNs),combining data-driven models with fundamental geotechnical principles to improve both the interpretability and predictive accuracy of AI applications in geotechnical engineering,marking a novel contribution to the field.

关键词： Laterally loaded drilled shaft load transfer and failure mechanisms Physics-informed neural networks(PINNs) P-y curves Artificial intelligence(AI) Dataset

来源：

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

Converting TRex-RGB green-channel data to 557.7 nm auroral intensity:Methodology and initial results

引用

earth and Planetary Physics 2024年第1期8卷 258-274页

作者： Jun Liang D.M.Gillies e.Spanswick e.F.Donovan University of Calgary CalgaryT2N 1N4Canada

The recently deployed Transition Region explorer(TRex)-RGB(red-green-blue)all-sky imager(ASI)is designed to capture“true color”images of the aurora and *** the 557.7 nm green line is usually the brightest emission line in visible auroras,the green channel of a TRex-RGB camera is usually dominated by the 557.7 nm *** this rationale,the TRex mission does not include a specific 557.7 nm imager and is designed to use the RGB green-channel data as a proxy for the 557.7 nm *** this study,we present an initial effort to establish the conversion ratio or formula linking the RGB green-channel data to the absolute intensity of 557.7 nm auroras,which is crucial for quantitative uses of the RGB *** illustrate two approaches:(1)through a comparison with the collocated measurement of green-line auroras from the TRex spectrograph,and(2)through a comparison with the modeled green-line intensity according to realistic electron precipitation flux measurements from low-earth-orbit satellites,with the aid of an auroral transport *** demonstrate the procedures and provide initial results for the TRex-RGB ASIs at the Rabbit Lake and Lucky Lake *** RGB response is found to be *** conversion ratios or formulas between RGB green-channel data and the green-line auroral intensity are given and can be applied immediately by TRex-RGB data *** methodology established in this study will also be applicable to the upcoming SMILe ASI mission,which will adopt a similar RGB camera system in its deployment.

关键词： TRex RGB green-line aurora auroral transport model

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：