检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Zhao, Zhilin Cao, Longbing Wang, Chang-Dong The Data Science Lab School of Computing and DataX Research Centre Macquarie University SydneyNSW2109 Australia The School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Guangdong Province Key Laboratory of Computational Science Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China

The integrity of training data, even when annotated by experts, is far from guaranteed, especially for non-IID datasets comprising both in- and out-of-distribution samples. In an ideal scenario, the majority of samples would be in-distribution, while samples that deviate semantically would be identified as out-of-distribution and excluded during the annotation process. However, experts may erroneously classify these out-of-distribution samples as in-distribution, assigning them lab.ls that are inherently unreliable. This mixture of unreliable lab.ls and varied data types makes the task of learning robust neural networks notably challenging. We observe that both in- and out-of-distribution samples can almost invariably be ruled out from belonging to certain classes, aside from those corresponding to unreliable ground-truth lab.ls. This opens the possibility of utilizing reliable complementary lab.ls that indicate the classes to which a sample does not belong. Guided by this insight, we introduce a novel approach, termed Gray Learning (GL), which leverages both ground-truth and complementary lab.ls. Crucially, GL adaptively adjusts the loss weights for these two lab.l types based on prediction confidence levels. By grounding our approach in statistical learning theory, we derive bounds for the generalization error, demonstrating that GL achieves tight constraints even in non-IID settings. Extensive experimental evaluations reveal that our method significantly outperforms alternative approaches grounded in robust statistics. Copyright © 2022, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Deep and Localized Transcranial Magnetic Stimulation in Rat Using Conventional Round Coil

Deep and Localized Transcranial Magnetic Stimulation in Rat ...

引用

General Assembly and Scientific Symposium, URSI

作者： Mai. Lu Shoogo Ueno Key Lab. of Opt-Electronic Technology and Intelligent Control of Ministry of Education Lanzhou Jiaotong University Lanzhou Gansu China Graduate School of Engineering Kyushu University Fukuoka Japan

Transcranial magnetic stimulation (TMS) shows significant values in both brain research and therapeutic applications in the treatment of various psychiatric and neurological disorders. Animal studies of TMS plays an important role in understanding of the cellular and molecular mechanisms underlying TMS induced effects in the brain. This work investigated TMS in rat brain by employing conventional round coil with varied coil positions. Three dimensional distributions of induced electric field in realistic rat model were calculated by 3-D impedance method. Simulations results indicated that the conventional round coil can be applied to rat TMS with improved focality while also keeping high stimulation intensities in deep brain regions.

关键词： Neurological diseases Solid modeling Impedance measurement Transcranial magnetic stimulation Simulation Rats Brain modeling

来源：评论

学校读者我要写书评

暂无评论

Learning embedded lab.l-specific features for partial multi-lab.l learning

引用

Pattern Recognition 2026年 170卷

作者： Xiaohan Xu Hao Wang Jialu Yao Zan Zhang School of Computer Science and Information Engineering Hefei University of Technology Hefei 230009 Anhui China Key Laboratory of Knowledge Engineering with Big Data (the Ministry of Education of China) Hefei University of Technology China Department of Electrical and Systems Engineering University of Pennsylvania Philadelphia 19104-6314 PA United States

Partial multi-lab.l learning (PML) aims to learn from instances with weak supervision, where each instance is associated with a set of candidate lab.ls, among which only a subset is valid. Most existing approaches rely on identical feature representations to distinguish all class lab.ls, overlooking the inherent distinctiveness of different lab.ls, which leads to suboptimal model performance. Although recent studies have attempted to address this limitation by tailoring lab.l-specific features, critical shortcomings remain: (1) isolated processing of feature tailoring and lab.l disambiguation fails to leverage their synergistic relationship, and (2) direct extraction of lab.l-specific features from the original feature space tends to yield unreliable results due to inherent noise and disturbances. This paper proposes a unified PML framework that jointly performs lab.l disambiguation, embedded lab.l-specific feature learning, and model induction. Within this framework, identifying ground-truth lab.ls and generating lab.l-specific features mutually reinforce each other, leading to continuous refinement. By customizing features from a compact and noise-free embedded space, the framework further ensures robustness and reliability in learning. Specifically, low-rank and sparse decomposition is employed to separate ground-truth lab.ls from noisy ones, while a linear embedding discriminant model simultaneously generates embedded lab.l-specific features and induces the model. Moreover, we enhance the classifier’s accuracy by assuming that the input and output spaces share local geometric structures, encouraging similar instances to have similar lab.l sets. Extensive experiments on sixty-six real-world and synthetic datasets demonstrate that the proposed approach significantly outperforms state-of-the-art methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Design of High Temperature Anticorrosion Diagnosis System for Atmospheric and Vacuum Distillation Unit Based on Forcecontrol and SQL Sever2014 5

Design of High Temperature Anticorrosion Diagnosis System fo...

引用

5th International Conference on Automation, Control and Robotics engineering, CACRE 2020

作者： Zhao, Liang Zhang, Zexin Wang, Lida Liu, Guichang Dalian University of Technology School of Control Science and Engineering Key Lab. of Intelligent Control and Optimization for Industrial Equipment of Ministry of Education Dalian China Dalian University of Technology School of Control Science and Engineering Dalian China Dalian University of Technology School of Chemical Engineering Dalian China

ISBN: (纸本)9781728198880

Due to the complex composition of the oil flowing through the refining unit, the equipment anticorrosion work has always been a major problem in petrochemical enterprises, and the corrosion problem is a key factor related to shortening the service life of the atmospheric and vacuum distillation unit. Aiming at the problem of high temperature corrosion of atmospheric and vacuum unit in a petrochemical enterprise and the current situation of corrosion management in the enterprise, this paper designs a high temperature corrosion diagnosis system of atmospheric and vacuum unit based on Forcecontrol configuration development platform. The system uses three dimensional interpolation algorithm to calculate the corrosion rate, and uses Microsoft SQL sever2014 database technology and ForceControl 7.1 configuration development platform to realize the comprehensive management of corrosion data, corrosion rate calculation, early warning, expert advice and other functions. Through the function test of each module in the system, and using the actual corrosion data provided by the enterprise to simulate and verify the system, the test results show that the system can achieve the expected functions, and ensure the calculation accuracy of corrosion rate. © 2020 IEEE.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

From Questions to Clinical Recommendations: Large Language Models Driving Evidence-Based Clinical Decision Making

arXiv

引用

arXiv 2025年

作者： Li, Dubai Jiang, Nan Huang, Kangping Tu, Ruiqi Ouyang, Shuyu Yu, Huayu Qiao, Lin Yu, Chen Zhou, Tianshu Tong, Danyang Wang, Qian Li, Mengtao Zeng, Xiaofeng Tian, Yu Tian, Xinping Li, Jingsong College of Biomedical Engineering and Instrument Science Zhejiang University Engineering Research Center of EMR and Intelligent Expert System Ministry of Education Hangzhou310027 China Ministry of Science and Technology State Key Laboratory of Complex Severe and Rare Diseases Key Laboratory of Rheumatology and Clinical Immunology Ministry of Education Beijing100730 China Research Center for Scientific Data Hub Zhejiang Lab Hangzhou311121 China

Clinical evidence, derived from rigorous research and data analysis, provides healthcare professionals with reliable scientific foundations for informed decision-making. Integrating clinical evidence into real-time practice is challenging due to the enormous workload, complex professional processes, and time constraints. This highlights the need for tools that automate evidence synthesis to support more efficient and accurate decision making in clinical settings. This study introduces Quicker, an evidence-based clinical decision support system powered by large language models (LLMs), designed to automate evidence synthesis and generate clinical recommendations modeled after standard clinical guideline development processes. Quicker implements a fully automated chain that covers all phases, from questions to clinical recommendations, and further enables customized decision-making through integrated tools and interactive user interfaces. To evaluate Quicker’s capabilities, we developed the Q2CRBench-3 benchmark dataset, based on clinical guideline development records for three different diseases. Experimental results highlighted Quicker’s strong performance, with fine-grained question decomposition tailored to user preferences, retrieval sensitivities comparable to human experts, and literature screening performance approaching comprehensive inclusion of relevant studies. In addition, Quicker-assisted evidence assessment effectively supported human reviewers, while Quicker’s recommendations were more comprehensive and logically coherent than those of clinicians. In system-level testing, collab.ration between a single reviewer and Quicker reduced the time required for recommendation development to 20-40 minutes. In general, our findings affirm the potential of Quicker to help physicians make quicker and more reliable evidence-based clinical decisions. © 2025, CC BY.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

knowledge Graph for China’s Genealogy

Knowledge Graph for China’s Genealogy

引用

IEEE International Conference on Big knowledge (ICBK)

作者： Xindong Wu Tingting Jiang Yi Zhu Chenyang Bu Key Laboratory of Knowledge Engineering with Big Data (Hefei University of Technology) Ministry of Education China School of Information Engineering Yangzhou University Yangzhou China

ISBN: (数字)9781728181561

ISBN: (纸本)9781728181578

Genealogical knowledge graphs depict the relationships of family networks and the development of family histories. They can help researchers to analyze and understand genealogical data, search for genealogical roots, and explore the origins of a family more easily. However, the multi-type, multisource dynamic changes and specialized nature of genealogical data bring challenges to the development of contemporary knowledge graph models. Applying existing methods to genealogical data can result in problems of overlooking certain specialized vocabulary and dynamic properties such as personal experiences. In this paper, we propose a genealogical knowledge graph model GKGM that combines HAO intelligence (h uman intelligence + a rtificial intelligence + o rganizational intelligence) and ontology granularity division technology to address the above problems. Furthermore, a method of applying the model to construct genealogical knowledge graphs is demonstrated, and an experiment conducted on a real-world genealogical dataset verifies the feasibility and effectiveness of the model.

关键词： Ontologies data models Blood data mining Biological system modeling History

来源：评论

学校读者我要写书评

暂无评论

A simple elastoplastic dynamic constitutive model of saturated structural soft clay

A simple elastoplastic dynamic constitutive model of saturat...

引用

China Rock 2020 Conference

作者： Feng, Shuangxi Lei, Huayang Qi, Ziyang Department of Civil Engineering Tianjin University Tianjin300354 China Key Laboratory of Coast Civil Structure Safety of Education Ministry Tianjin University Tianjin300354 China Key Lab. of Comprehensive Simulation of Engineering Earthquake and Urban-Rural Seismic Resilience Cea Tianjin300354 China

The dynamic constitutive relation of soil is widely applied to the ground design of marine structure and infrastructure in offshore geotechnical engineering. Increasing evidence from marine geotechnical design has shown that the traditional dynamic constitutive model has uncertain theories, many parameters, and complexity. Therefore, it is urgent to overcome these limitations and develop a simple and practical elastoplastic dynamic constitutive relation for the saturated structural soft clay. A simple dynamic constitutive relation with 5 parameters was proposed in this study, which can reflect the strain accumulation, hysteretic characteristics and cyclic degradation of the stress-strain for structural soft clay. First, the Drucker-Prager model was selected as the bounding surface, hardening modulus equation was established by the hardening law, interpolation function and structural strength coefficient. Then, the incremental stress-strain relationship was derived based on the incremental theory. The UMAT subroutine was compiled by FORTRAN language. Finally, the cyclic triaxial test and numerical simulation were carried out, and the stress-strain relationship was analyzed. The research results showed that the test results were in good agreement with the prediction results of the model. © Published under licence by IOP Publishing Ltd.

关键词： Constitutive models

来源：评论

学校读者我要写书评

暂无评论

Advances in Tumor Targeting Biomimetic Drug Delivery Systems: A Promising Approach for Antitumor Therapy

Pharmaceutical Fronts

引用

Pharmaceutical Fronts 2024年第2期6卷 E101-E118页

作者： Mo, Ziyi He, Jiao Li, Man Guo, Rong He, Qin Key Lab. of Drug-Targeting and Drug Delivery System of the Education Ministry and Sichuan Province Sichuan Engineering Laboratory for Plant-Sourced Drug Sichuan Research Center for Drug Precision Industrial Technology West China School of Pharmacy Med-X Center for Materials Sichuan University Department of Biochemistry and Molecular Biology West China School of Basic Medical Sciences and Forensic Medicine Sichuan University

Cancer is one of the most fatal diseases that attract numerous efforts and attention from researchers. Among plentiful therapeutic agents, chemotherapy is frequently used in treating virulent tumors, and its insistent administration is useful in the ablation of cancers;however, it also produces side effects. Biomimetic drug delivery systems (BDDSs) provide an alternative route for antitumor therapy. Their endogenous substances may be extracellular vesicles, living cells, cell membranes, etc., which optimize single-agent chemotherapy. They upgrade traditional drug delivery platforms by combining the original drug with itself, disguised as a Trojan Horse, to trick the immune system or tumor tissues to achieve higher targeting and lower immunogenicity. Herein, we review three BDDS strategies being used recently in antitumor drug development and their advances, aiming at providing general guidelines and opportunities in this field in the future. © 2024. The Author(s).

关键词： antitumor biomimetic drug delivery system endogenous substances nanoparticles

来源：评论

学校读者我要写书评

暂无评论

Improving Fracture Toughness of Tc11/Tc17 Linear Friction Weld Via Solution Aging Heat Treatment

SSRN

引用

SSRN 2023年

作者： Zhao, Pengkang Hu, Yingxin Chu, Qiaoling Geng, Peihao Yuan, Liang Cao, Fujun Li, Shujuan Tao, Jun Ma, Ninshu Key Lab. of CNC Machine Tool and Machinery Manufacturing Equipment Integration the Ministry of Education Xi’an University of Technology Xi’an710048 China Joining and Welding Research Institute Osaka University Osaka567-0047 Japan School of Materials Science and Engineering Xi’an University of Technology Xi’an710048 China College of Bioresources Chemical and Materials Engineering Shaanxi University of Science and Technology Xi′an710021 China Nanjing211106 China Aeronautical Key Laboratory for Welding and Joining Technologies AVIC Manufacturing Technology Institute Beijing100024 China

The titanium alloy's linear friction welding (LFW) joints exhibit good strength and fatigue properties. However, their limited toughness hampers the widespread application of LFW technology. This article explores how to improve the toughness of TC11/TC17 LFW joints through the solution aging method and the influence mechanism of dislocation density and crystal texture on its improvement. The research indicates that the fracture toughness of the joint significantly increases from 38% to 88% of the TC11 base material (BM) after solution and high-temperature (HT) aging. The original joint exhibits brilliant crystalline facets and limited dimples, while the fracture surface appears rough with parallel fatigue stripes and numerous uniform dimples after solution and HT aging. Dislocations are widely dispersed in the α phase in TC11 WZ, and the dislocation wall composed of interlaced dislocation lines appears in TC17 WZ. However, the dislocation defects and density decrease significantly after solution aging, and the increased aging temperature promotes the precipitation of refined and coherent secondary phases. Grain orientation becomes more random in three directions, with the proportion of high Schmid factor (SF) values (μ>0.3) increasing from 70.9% to 88.1%. The hardness values in the joint become more uniform, significantly reducing the disparity on both sides with increasing aging temperature. The research sheds light on the evolution of dislocation and crystal texture under solution aging conditions, clarifying the internal mechanism behind the joint's improved toughness. © 2023, The Authors. All rights reserved.

关键词： Fracture toughness

来源：评论

学校读者我要写书评

暂无评论

Research on combustion properties and thermokinetic of pulverized coal: Raw and pre-oxidized samples 4

Research on combustion properties and thermokinetic of pulve...

引用

2020 4th International Workshop on Advances in Energy Science and Environment engineering, AESEE 2020

作者： Zhai, Xiao-Wei Zhai, Xiao-Wei Li, Jie Li, Jie Song, Bobo Song, Bobo School of Safety Science and Engineering Xi'An University of Science and Technology Xi'an Shaanxi710054 China Key Lab. of Ministry of Education for Prevention and Control of Mining Disasters in Western China Xi'an Shaanxi710054 China

The combustion behaviors of raw and pre-oxidized samples for two pulverized coals (PC) were investigated using a thermogravimetric (TG) analyzer. The mass variations, characteristic temperatures and combustion characteristic parameters, et al. of PC during oxidation combustion were analyzed. The results demonstrated that the ignition temperatures (T i) and burnout temperatures (T b) of pre-oxidized PC were lower than those of raw PC. The total mass losses of the four samples were 91.33, 92.32, 88.85, and 90.02 mass% respectively, which were consistent with the volatile matter and fixed carbon contents (Vad+FCad). The calculated combustion characteristic parameters (the ignition index C i, burnout index C b, and comprehensive combustibility index S n) of pre-oxidized samples were all greater than those of raw samples, which indicated that pre-oxidized PC was more prone to self-ignition than raw PC under the same conditions. The thermokinetic parameters of PC during pyrolysis and combustion were studied by Coats-Redfern method. The apparent activation energy E a of these four PC samples ranged from 45.12 to 55.37 kJ/mol. © Published under licence by IOP Publishing Ltd.

关键词： Oxidation

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：