检索结果-内蒙古大学图书馆

Layer-Wise Learning Rate Optimization for Task-Dependent Fine-Tuning of Pre-Trained Models: An Evolutionary Approach

引用

ACM Transactions on Evolutionary Learning and Optimization 2024年第4期4卷 1-23页

作者： Bu, Chenyang Liu, Yuxin Huang, Manzong Shao, Jianxuan Ji, Shengwei Luo, Wenjian Wu, Xindong Key Laboratory of Knowledge Engineering with Big Data Ministry of Education and School of Computer Science and Information Engineering Hefei University of Technology Hefei China School of Artificial Intelligence and Big Data Hefei University Hefei China Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies School of Computer Science and Technology Harbin Institute of Technology Shenzhen China

The superior performance of large-scale pre-Trained models, such as Bidirectional Encoder Representations from Transformers (BERT) and Generative Pre-Trained Transformer (GPT), has received increasing attention in both academic and industrial research and has become one of the current research hotspots. A pre-Trained model refers to a model trained on large-scale unlabeled data, whose purpose is to learn general language representation or features for fine-Tuning or transfer learning in subsequent tasks. After pre-Training is complete, a small amount of labeled data can be used to fine-Tune the model for a specific task or domain. This two-stage method of "pre-Training+fine-Tuning"has achieved advanced results in natural language processing (NLP) tasks. Despite widespread adoption, existing fixed fine-Tuning schemes that adapt well to one NLP task may perform inconsistently on other NLP tasks given that different tasks have different latent semantic structures. In this article, we explore the effectiveness of automatic fine-Tuning pattern search for layer-wise learning rates from an evolutionary optimization perspective. Our goal is to use evolutionary algorithms to search for better task-dependent fine-Tuning patterns for specific NLP tasks than typical fixed fine-Tuning patterns. Experimental results on two real-world language benchmarks and three advanced pre-Training language models show the effectiveness and generality of the proposed framework. © 2024 held by the owner/author(s).

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Boosting spring runoff into the sea by reservoir regulation and its potential for estuarine fishery recovery

引用

science China Earth sciences 2024年第5期67卷 1591-1603页

作者： Xiao WU Houjie WANG Yongyong FAN Naishuang BI Chiyuan MIAO Jingping XU Zuosheng YANG College of Marine Geosciences Key Laboratory of Submarine Geosciences and Prospecting Technologyand Frontiers Science Center for Deep Ocean Multispheres and Earth SystemOcean University of ChinaQingdao 266100China Laoshan National Laboratory Qingdao 266061China Institute of the Yellow River Estuary and Coast Science Dongying 257000China State Key Laboratory of Earth Surface Processes and Resource Ecology Faculty of Geographical ScienceBeijing Normal UniversityBeijing 100875China Department of Ocean Science&Engineering Southern University of Science and TechnologyShenzhen 518055China

With global surge in reservoir construction over the past decades,river systems worldwide have been profoundly ***,flow manipulation by reservoirs has altered the natural hydrological processes,resulting in extensive modifications of fluvial-marine *** the adverse ecological consequences of reservoirs has become a global concern and has garnered increasing *** Yellow River,as one of the most extensively manipulated river systems globally,has experienced substantial changes in the amount and timing of water discharge due to the presence of numerous reservoirs scattered throughout its catchment *** alterations have caused physicochemical changes in the estuary and subsequent modifications to the estuarine *** recent years,the Yellow River Conservancy Committee initiated the release of water through the Xiaolangdi Dam during the major spawning period of fisheries,specifically in the spring,with the aim of improving the estuarine ecological *** 2011 to 2020,a total of 84.05 km^(3)of water was discharged from the Xiaolangdi Reservoir during spring seasons,of which 40%(33.16 km^(3))constituted water impounded within the reservoir during preceding ***,the spring water discharge from the Yellow River to the sea increased significantly from 1.50 km^(3)/yr to 3.46 km^(3)/yr in the past decade,leading to a decrease in estuarine salinity by 1.6 *** estuarine fishery resources,such as fish eggs in the Yellow River estuary,have demonstrated evident *** reservoir regulation in the Yellow River,which has successfully enhanced spring water discharge and subsequently restored estuarine fishery resources,presents an effective attempt for mitigating the adverse ecological effects associated with reservoirs.

关键词： Reservoir regulation Water release Spring runoff Estuarine ecosystem Yellow River

来源：评论

学校读者我要写书评

暂无评论

H∞ fault-tolerant fuzzy intermittent control for nonlinear hyperbolic PDE systems with multiple delays and actuator failures

引用

ISA Transactions 2025年 156卷 39-60页

作者： Zhang, Xu Gao, Xuan Wang, Zi-Peng Luo, Biao Han, Ke-Zhen Yan, Xue-Hua School of Automation Central South University Changsha410083 China School of Electrical Engineering University of Jinan Jinan250022 China School of Information Science and Technology Beijing Laboratory of Smart Environmental Protection Beijing Key Laboratory of Computational Intelligence and Intelligent System and Beijing Institute of Artificial Intelligence Beijing University of Technology Beijing100124 China

This study presents an H∞ fault-tolerant fuzzy intermittent control approach for the nonlinear hyperbolic partial differential equation (PDE) systems with multiple delays and actuator failures (MDAFs). Firstly, the nonlinear hyperbolic PDE systems with MDAFs are characterized by the Takagi–Sugeno (T–S) fuzzy delayed hyperbolic PDE model. Next, by employing the Lyapunov direct method, this paper demonstrates the robust exponential stability using spatial linear matrix inequalities (SLMIs) based on a new switching Lyapunov functional (LF). Furthermore, the H∞ fault-tolerant fuzzy intermittent control issue for nonlinear hyperbolic PDE systems with MDAFs is transformed into the LMI feasibility problem to deal with the SLMIs. Lastly, the feasibility of the constructed control strategy is demonstrated by two illustrative examples. © 2024 International Society of Automation

关键词： Lyapunov functions

来源：评论

学校读者我要写书评

暂无评论

Low frequency error analysis and calibration for multiple star sensors system of GaoFen7 satellite

引用

地球空间信息科学学报（英文版） 2024年第1期27卷 82-94页

作者： Yanli Wang Mi Wang Ying Zhu Xiaoxiang Long The College of Geodesy and Geomatics Shandong University of Science and TechnologyQingdaoChina The State Key Laboratory of Information Engineering in Surveying Mappingand Remote SensingWuhan UniversityWuhanChina School of Electrical and Information Engineering Wuhan Institute of TechnologyWuhanChina Value-added Service Department China Center for Resources Satellite Data and ApplicationBeijingChina

The GaoFen7(GF7)optical satellite is the first Chinese civilian sub-meter high-resolution stereo mapping satellite and is equipped with a double linear array camera and laser altimeter to achieve large-scale topographic *** improve the accuracy of attitude determination,an attitude determination system comprised of four star sensors is *** to the measurement accuracy and steady performance,the star sensors 1a and 1b is usually used together for satellite attitude calculation,which is called the conventional mode of attitude ***,the combination of star sensors 2a and 2b is called the unconventional mode of attitude *** by variations in the incident angle of sunlight and solar radiation,thermal deformation occurs in the body and installation structure of the star sensor,which causes Attitude Low-Frequency Error(ALFE)and seriously influences the con-sistency of attitude determination results of different combination modes for multiple star sensors *** study proposes an ALFE analysis and calibration approach for the multiple star sensors system of GF7 satellite to ensure the consistency of attitude determination results of different combination *** on the statistical characteristics of the angles of the three axes,the installation parameters of the four star sensors are first *** analyzing the characteristics of the optical axis angles within 1420 orbit periods over 135 days,the segmented ALFE compensation model between the unconventional and conven-tional modes is proposed based on the Fourier series model and input parameter of *** on the on-orbit installation parameters and the ALFE model,the precise attitude determination results of the unconventional mode are *** results show that the attitude determination consistency after compensation is better than 2".Moreover,the reliable application time range of the compensation model is 30 days to satisfy the re

关键词： Attitude Low Frequency Error(ALFE) multiple star sensors system GaoFen7 attitude determination consistency Fourier series

来源：评论

学校读者我要写书评

暂无评论

Event-Triggered Cooperative Tracking Control of Multiagent systems with a Dynamic Leader via Approximate Dynamic Programming

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第6期5卷 2752-2765页

作者： Fu, Hao He, Haodong Chen, Yang Wuhan University of Science and Technology Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System The School of Computer Science and Technology Wuhan430081 China Wuhan University of Science and Technology Eng. Res. Center for Metallurgical Automation and Measurement Technology of Ministry of Education Wuhan430081 China

This article investigates the optimal cooperative tracking control problem for unknown nonlinear multiagent systems with a real dynamic leader and uncertainties. Its main difficulty lies in eliminating effect of the unmatched uncertainty and noncooperation in the event-triggered manner. To overcome this difficulty, a distributed robust event-triggered control algorithm is developed with the aid of a disturbance observer and robust approximate dynamic programming (ADP). Specifically, this problem is converted into a cooperative multiplayer game for the distributed auxiliary systems employing observer-based feedforward compensation, which contributes to reducing conservatism;via an adaptive event-triggered mechanism and a new weight tuning law, a distributed robust ADP approach is proposed with hope to obviate judging the existence of the saddle point beforehand. Moreover, the local neighbor tracking error and the weight estimation error are demonstrated to be uniformly ultimately bounded by means of Lyapunov stability theory. We also prove that Zeno behavior is excluded. A comparative simulation example for multiquadrotor systems is carried out to verify the effectiveness and superiority of our algorithm. © 2020 IEEE.

关键词： Dynamic programming

来源：评论

学校读者我要写书评

暂无评论

SSD object detection algorithm based on knowledge map

引用

International Journal of Wireless and Mobile Computing 2024年第2期27卷 152-160页

作者： Huang, Li Wang, Xiaofeng Lu, Jianhua Hu, Wei Zhang, Changrong College of Computer Science and Technology Wuhan University of Science and Technology Hubei Wuhan China China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan University of Science and Technology Hubei Wuhan China Key Laboratory of Metallurgical Equipment and Control Technology of the Ministry of Education Wuhan University of Science and Technology Hubei Wuhan China Precision Manufacturing Research Institute Research Centre for Biomimetic Robot and Intelligent Measurement and Control Wuhan University of Science and Technology Hubei Wuhan China

With the pervasive integration of artificial intelligence into all aspects of human life, talent emerges as a primary resource. Upon analysing the current state of talent training in higher education institutions, issues such as dispersed knowledge points, overlapping content, a singular practice approach and ineffective evaluation have been identified. In response to these challenges, this paper proposes a multidisciplinary and comprehensive practical teaching methodology grounded in the knowledge graph framework. It delves into diverse paths for practical teaching and assessment, including aspects like teaching objectives, problem decomposition, resource integration, implementation methods and performance evaluation. The practical application of the SSD algorithm in researching service robot indoor object detection serves as an illustrative example. Employing the holistic practical teaching approach facilitated by the knowledge graph, this model guides students in acquiring object detection expertise, thereby enhancing their comprehensive development. Copyright © 2024 Inderscience Enterprises Ltd.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Distinct MLT asymmetry of auroral kilometric radiation observed by the FAST satellite

引用

Earth and Planetary Physics 2025年第1期9卷 188-193页

作者： JiaWen Tang Sai Zhang FuLiang Xiao HongMing Yang Si Liu YiHua He Chang Yang YuYue Jin ZhouKun Deng Ping Li Archie James Johnston School of Physics and Electronic Sciences Changsha University of Science and TechnologyChangsha 410114China Hunan Province Higher Education Key Laboratory of Modeling and Monitoring on the Near-Earth Electromagnetic Environments Changsha University of Science and TechnologyChangsha 410114China School of Electrical and Information Engineering Changsha University of Science and TechnologyChangsha 410114China

Auroral Kilometric Radiation (AKR) is a common radio emission,which can contribute to the magnetosphere-ionosphereatmosphere co u *** emissions have been observed in all magnetic planet magnetospheres of the solar *** this study,using observations from the FAST satellite from 30 August 1996 to 9 September 2001,the distribution of AKR in altitude=500-4500 km and invariant latitude (|ILAT|)=60°-80°has been analyzed.63045 AKR samples have been identified with~48%(52%) samples on the dayside (nightside).Of considerable interest,there is a distinct MLT asymmetry with the high occurrence rate in MLT=05-08 and 18-22(02-05 and 12-17) in the northern (southern) *** distinct MLT asymmetry is associated with the direction of Bxof the interplaneta ry magnetic *** addition,the occurrence rate on the nightside clearly increases as the AE^(*) index *** study further enriches the information and understanding of AKR in the magnetosphere as well as other similar radio emissions.

关键词： asymmetry auroral kilometric radiation interplanetary magnetic field

来源：评论

学校读者我要写书评

暂无评论

Scalable graphene foam with ultrahigh conductivity for stabilizing Pt towards efficient hydrogen evolution

引用

Nano Research 2024年第8期17卷 6968-6976页

作者： Ming Zhao Meng Huang Huihui Jin Dayin He Wei Qian Zixin Zhang Daping He State Key Laboratory of Advanced Technology for Materials Synthesis and Processing Wuhan University of TechnologyWuhan 430070China Hubei Engineering Research Center of RF-Microwave Technology and Application Wuhan University of TechnologyWuhan 430070China School of Information Engineering Wuhan University of TechnologyWuhan 430070China School of Chemistry and Materials Science University of Science and Technology of ChinaHefei 230026China Sanya Science and Education Innovation Park of Wuhan University of Technology Sanya 572000China

For the carbon-based catalyst to be active and stable,especially in harsh electrochemical environments,the key is to decrease the concentration of defects and raise the degree of graphitization of the carbon ***,we develop a highly graphitized graphene foam with multiplicated structure to fabricate self-supporting Pt-based catalysts for efficient and stable hydrogen evolution reaction(HER)*** foam(GO-2850)is obtained through an ultra-high temperature treatment at 2850℃,with perfect graphene structure and extremely low defect,ensuring high electrical conductivity and corrosion ***,its multiplicated structure provides an inherently favorable environment for the dispersion of Pt nanoparticles(Pt NPs)and offers abundant channels for electrolyte infiltration during the catalytic *** a result,the as-prepared Pt/GO-2850 is far active and stable than the Pt NPs supported on commercial carbon paper(Pt/CP)counterpart toward catalyzing HER,exhibiting an outstanding activity and long-term durability(300 h@10 mA·cm^(−2))in acidic/alkaline/seawater *** can be attributed to the stronger interaction between the lower-defect GO-2850 substrate and Pt,as evidenced by characterization and theoretical *** work extends further insight into the design self-supporting catalysts of high activity and stability with promising prominent application toward green energy devices.

关键词： graphene foam high graphitization Pt nanoparticles hydrogen evolution reaction long-term stability

来源：评论

学校读者我要写书评

暂无评论

Effect of Testing Temperature Gradients on Dielectric Polarization Parameters of Transformer Oil-paper Insulation

引用

IEEE Transactions on Dielectrics and Electrical Insulation 2024年第3期32卷 1460-1467页

作者： Zhou, Jiacheng Liu, Jiefeng Jiang, Zaijun Gong, Heyan Fan, Xianhao Lin, Yongqing Wang, Qian Gao, Bing Guangxi University Guangxi Key Laboratory of Power System O Ptimization and Energy Technology China Tsinghua University Department of Electrical Engineering Beijing100084 China Hunan University College of Electrical and Information Engineering Changsha410082 China

Frequency domain spectroscopy (FDS) plays an important role in analyzing the aging state of power transformer oil-paper insulation and ensuring the stability of power systems. However, in existing research, the oil-paper insulation's FDS is collected under a uniform testing temperature field, overlooking the impact of testing temperature gradients existing in actual transformers. To address this issue, the FDS of oil-paper insulation samples is measured under various temperature gradients. Subsequently, logarithmic-derivative spectroscopy (LDS) is obtained from the measured FDS, and an improved equation is proposed to fit the LDS and extract four polarization parameters related to the oil-paper insulation's aging state. Furthermore, the changing patterns of the above parameters are analyzed at a microscopic scale based on the dielectric relaxation theory, and the quantitative equations between each polarization parameter and testing temperature gradients are established. To validate the accuracy of established equations, new samples are prepared and the calculated relative error in each parameter is less than 15%. Finally, this research provides a potential method for eliminating the impact of testing temperature gradients on oil-paper insulation's FDS. © 2024 IEEE.

关键词： Oil filled transformers

来源：评论

学校读者我要写书评

暂无评论

Long cycle performance of NC@VN/MnO cathode for AZIBs based on Mn/V relay type collaboration

引用

Journal of Energy Chemistry 2023年第8期83卷 106-118,I0005页

作者： Tingting Li Ruisong Guo Yang Li Leichao Meng Xiaohong Sun Fuyun Li Xinqi Zhao Zhongkai Xu Jianhong Peng Lingyun An Key Laboratory of Advanced Ceramics and Machining Technology of Ministry of Education School of Materials Science and EngineeringTianjin UniversityTianjin 300072China College of Petrochemical Technology Lanzhou University of TechnologyLanzhou 730050GansuChina Qinghai Provincial Key Laboratory of Nanomaterials and Technology School of Physics and Electronic Information EngineeringQinghai Minzu UniversityXining 810007QinghaiChina

Aqueous zinc ion batteries(AZIBs) have received great attention because of their non-toxicity,high safety,low cost,high abundance,and high specific ***,their specific capacity is still low compared with lithium ion battery,and current academic research interesting has been focused on developing new cathode materials with high specific *** this study,a Mn/V hybrid polymer framework is designed by a simple self-polymerization *** subsequent calcination,ultrafine VN quantum dots and MnO nanoparticles are generated in situ and stably encapsulated inside N-doped carbon(NC) shells to obtain a novel hybrid cathode NC@VN/MnO for *** to the density functional theory(DFT) calculation,the hybrids of MnO and VN can generate both interfacial effects and built-in electric fields that significantly accelerate ion and electron transport by tuning the intrinsic electronic structure,thus enhancing electrochemical performance.A synergistic strategy of composition and structural design allows the rechargeable AZIBs to achieve low-cost and excellent long-cycle performance based on a relay type collaboration at different cycling ***,the NC@VN/MnO cathode has output a capacity of 108.3 mA h g^(-1)after 12,000 cycles at 10 A g^(-1).These results clearly and fully demonstrate the advantages of the hybrid cathode NC@VN/MnO.

关键词： VN quantum dots Mno N-doped carbon Heterojunction Long-cycle performance

来源：评论

学校读者我要写书评

暂无评论

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

请选择保存的检索档案：

请选择收藏分类：

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

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

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：