检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Mitsuka, Yoshihiro Golestan, Shadan Sufiyan, Zahin Schoepp, Sheila Miwa, Shotaro Zaïane, Osmar r. Information Technology R&D Center Mitsubishi Electric Corporation Japan Department of Computing Science University of Alberta Canada Advanced Technology R&D Center Mitsubishi Electric Corporation Japan Alberta Machine Intelligence Institute Canada

The scheme of adaptation via meta-learning is seen as an ingredient for solving the problem of data shortage or distribution shift in real-world applications, but it also brings the new risk of inappropriate updates of the model in the user environment, which increases the demand for explainability. Among the various types of XAI methods, establishing a method of explanation based on past experience in meta-learning requires special consideration due to its bi-level structure of training, which has been left unexplored. In this work, we propose influence functions for explaining meta-learning that measure the sensitivities of training tasks to adaptation and inference. We also argue that the approximation of the Hessian using the Gauss-Newton matrix resolves computational barriers peculiar to meta-learning. We demonstrate the adequacy of the method through experiments on task distinction and task distribution distinction using image classification tasks with MAML and Prototypical Network. © 2025, CC BY.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Anti-inflammatory and anabolic biphasic scaffold facilitates osteochondral tissue regeneration in osteoarthritic joints

引用

Journal of Materials Science & technology 2023年第25期156卷 20-31页

作者： Xiangbo Meng Ling Li Cuishan Huang Keda Shi Qingqiang Zeng Chunyi Wen Sibylle Grad Mauro Alini Ling Qin Xinluan Wang Translational Medicine R&D Center Institute of Biomedical and Health EngineeringShenzhen Institutes of Advanced TechnologyChinese Academy of SciencesShenzhen 518055China AO Research Institute Davos Clavadelerstrasse 8Davos Platz 7270Switzerland Musculoskeletal Research Laboratory Department of Orthopaedics & TraumatologyThe Chinese University of Hong KongHong Kong SAR 999077China CAS-HK Joint Lab of Biomaterials Joint Laboratory of Chinese Academic of Science and Hong Kong for BiomaterialsTranslational Medicine Research and Development CenterShenzhen Institutes of Advanced Technology of Chinese Academy of Sciences and The Chinese University of Hong KongChina Department of Biomedical Engineering Faculty of EngineeringThe Hong Kong Polytechnic UniversityHong Kong SAR 999077China

Osteochondral defects (OCd) are common but difficult to heal due to the low intrinsic repair capacity of cartilage and its complex hierarchical structure. In osteoarthritis (OA), OCd become more challenging to repair as both cartilage and subchondral bone regeneration are further impaired due to the arthritic environment. Numerous biomaterials have been developed and tested in osteochondral defects while ignoring the inflammatory environment. To target this challenging underlying pathophysiology, we designed and fabricated a biphasic porous and degradable scaffold incorporating anti-inflammatory and anabolic molecules by low-temperature rapid prototyping technology, and its effects on promoting osteochondral regeneration were evaluated using our well-established OA-OCd rabbit model. The biphasic porous scaffolds consisted of poly lactic-co-glycolic acid (PLGA) with kartogenin (KGN) for cartilage repair and PLGA and β-calcium phosphate (PLGA/β-TCP) with cinnamaldehyde (CIN) for subchondral bone repair. KGN is a molecule for promoting chondrogenesis and CIN is a phytomolecule for enhancing osteogenesis and alleviating inflammation. The biphasic scaffolds PLGA/KGN-PLGA/β-TCP/CIN (PK/PTC) with bio-mimic structure provided stable mechanical properties and exhibited excellent biocompatibility to support cell adhesion, proliferation, migration, and distribution. Furthermore, KGN and CIN within biphasic scaffolds could be released in a controlled and sustained mode, and the biphasic scaffold degraded slowly in vitro . Evaluating the repair of 16-weeks post-implantation into critically sized OA-OCd rabbit models revealed that the biphasic scaffold could promote subchondral bone and cartilage regeneration, as well as reverse subchondral osteosclerosis caused by inflammation in vivo . These findings support the utilization of the PK/PTC scaffold for osteochondral regeneration and provide a promising potential strategy for clinical application for the treatment of patients with

关键词： Biphasic scaffold Osteochondral defect Osteoarthritis Kartogenin Cinnamaldehyde Anti-inflammation

来源：评论

学校读者我要写书评

暂无评论

deformable micro-supercapacitor fabricated via laser ablation patterning of Graphene/liquid metal

引用

npj Flexible Electronics 2024年第1期8卷 770-778页

作者： Keon-Woo Kim Seong Ju Park Su-Jeong Park Inae Kim Bomi Park Se Hyun Kim Unyong Jeong Jin Kon Kim Chanwoo Yang National Creative Research Initiative Center for Hybrid Nano Materials by High-level Architectural Design of Block Copolymer Department of Chemical EngineeringPohang University of Science and Technology(POSTECH)PohangGyeongbuk 37763Republic of Korea Advanced Nano-Surface&Wearable Electronics Research Laboratory Heat and Surface Technology R&D DepartmentKorea Institute of Industrial TechnologyIncheon 21999Republic of Korea Division of Chemical Engineering Konkuk UniversitySeoul 05029Republic of Korea Department of Materials Science and Engineering Pohang University of Science and Technology(POSTECH)PohangGyeongbuk 37763Republic of Korea

deformable andminiaturized energy storage devices are essential for powering soft ***,we fabricate deformablemicro supercapacitors(MSCs)based on eutectic gallium-indium liquidmetal(EGaIn)current collectors with integrated *** well-define interdigitated electrode patterning with controlled gap is successfully realized by using the laser ablation because of a strong laser absorption of graphene and *** judicious control of gap size between neighboring interdigitated electrodes and mass loading of graphene,we achieve a high areal capacitance(1336μF cm^(−2))with reliable rate *** addition,owing to the intrinsic liquid characteristics of EGaIn current collector,the areal capacitance of fabricated MSC retains 90% of original value evenafter repetitive foldingand20%stretching up to 1000 ***,we successfully integrate deformable MSC with a commercial light-emitting diode to demonstrate the feasibility ofMSC as a deformable power *** fabricated MSCs operate stably under various mechanical deformations,including stretching,folding,twisting,and wrinkling.

关键词： laser stretching liquid

来源：评论

学校读者我要写书评

暂无评论

Exploring the Potential of deep Learning in Parkinson's disease drug discovery

Exploring the Potential of Deep Learning in Parkinson's Dise...

引用

2024 International Conference on Advances in Computing research on Science Engineering and technology, ACrOSET 2024

作者： Kumar, ravula Arun Ponnrajakumari, M. Kumar, M ravi Nirisha, A. Chandrasekaran, raja Moharekar, Tejashree Tejpal Vardhaman College of Engineering Department of Cse Telanagana India Velammal Engineering College Department of Ece Chennai India Kl University Department of Cse Vijayawada India Mlr Institute of Technology Department. of Csit Telangana India Sagunthala R&d Institute of Science and Technology Department of Ece Chennai India Yashwantrao Chavan School of Rural Development Shivaji University Kolhapur India

ISBN: (纸本)9798350388800

Complex and complex neurological illnesses like Parkinson's disease (Pd) are difficult to investigate. This publication introduces 'Neural Pathway Profiling for Parkinson's disease drug discovery.' It leverages cutting-edge deep learning. GCNs explain molecular interactions, LSTMs examine temporal sequences, and VAEs learn by forming new patterns. Our research uses genes, proteins, and clinical data to explain how Parkinson's disease's various components interact in complex ways. LSTMs illustrate how clinical data is connected to time, GCNs reveal how molecules interact in biological networks, and VAEs discover hidden chemical structures for generative learning. The blended design reveals key brain connections and Parkinson's disease therapeutic targets. © 2024 IEEE.

关键词： drug discovery

来源：评论

学校读者我要写书评

暂无评论

Accelerating deep Neural Networks with a Low-Cost Lossless Compression

Accelerating Deep Neural Networks with a Low-Cost Lossless C...

引用

International Conference on Electronics, Information and Communications (ICEIC)

作者： Seongwook Kim Gwangeun Byeon Seungeui Lee Youngseok Bae Seokin Hong Department of Electrical and Computer Engineering Sungkyunkwan University Suwon Korea Advanced R&D Center Hanwha Systems Co. Ltd. Emerging Science and Technology Directorate Agency for Defense Development

ISBN: (数字)9798331510756

ISBN: (纸本)9798331510763

Since deep learning inference involves a significant amount of computations, there have been a lot of efforts to accelerate the inference process by eliminating ineffectual compu-tations. As a solution to this problem, several sparsity-aware deep learning accelerators have been proposed. These accelerators perform the inference on compressed data (filters and feature maps) and dynamically remove the ineffectual multiplications with zero operands. For efficiently handling the compressed data on the accelerators, a simple and effective compression scheme, which encodes the whole data with its nonzero values and their positions, is necessary. However, prior approaches have several shortcomings, such as low compression efficiency and high complexity in compression and decompression hardware. This paper proposes a low-cost compression scheme that can be applied to sparsity-aware deep-learning accelerators. The compression scheme synergistically combines two low-cost compression schemes: Zero-Value Compression (ZVC) and Base-delta-Immediate Compression (BdI). Our experimental evaluation shows that the proposed compression scheme achieves an average compression ratio of 2.26, resulting in 2.33x performance improvement,

关键词： deep learning Filters Artificial neural networks Performance gain Complexity theory Hardware acceleration

来源：评论

学校读者我要写书评

暂无评论

Experimental Evaluation of deep Learning Assisted Autism Spectrum disorder Prediction Scheme using Convolutional Classification Principle 5

Experimental Evaluation of Deep Learning Assisted Autism Spe...

引用

5th International Conference on data Intelligence and Cognitive Informatics, ICdICI 2024

作者： Kumar, Abhinav Thapak, Sonika Thatipudi, Jayendra Gopal rameshkumar, C. Krishnamoorthy, r. Muralitharan, divya Ies College of Technology Department of Computer Science and Engineering Bhopal India University of North Texas Dept. of Computer Science United States Sathyabama Institute of Science and Technology Department Physics India Chennai Institute of Technology Centre for Advanced Wireless Integrated Technology Chennai India Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Department of C.S.E Chennai India

ISBN: (纸本)9798350389609

Autism Spectrum disorder (ASd) is characterized by challenges in social communication and restricted or repetitive behaviors. With rising ASd incidence, early diagnosis and intervention become increasingly crucial for improving developmental trajectories and quality of life. Traditional diagnostic methods are often subjective and limited. This study evaluates a deep learning-based approach for ASd prediction using Convolutional Neural Networks (CNNs) and recurrent Neural Networks (rNNs). Leveraging the Autism Brain Imaging data Exchange (ABIdE) dataset, which includes neuroimaging, genetic, and behavioral data, the model integrates spatial and temporal features to enhance prediction accuracy. The combined feature set yielded the highest accuracy of 93.7%, demonstrating the efficacy of integrating multiple data types. The proposed Convolutional recurrent Hybrid (CrH) model achieved a superior accuracy of 98%, outperforming traditional models. This research highlights the potential of deep learning techniques in improving ASd diagnostics. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Negative data in data Sets for Machine Learning Training

引用

OrGANIC LETTErS 2023年第17期25卷 2945-2947页

作者： Maloney, Michael P. Coley, Connor W. Genheden, Samuel Carson, Nessa Helquist, Paul Norrby, Per-Ola Wiest, Olaf Department of Chemistry and Biochemistry University of Notre Dame Notre Dame Indiana 46556 United States Department of Chemical Engineering and Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology Cambridge Massachusetts 02139 United States Molecular AI Discovery Sciences R&D AstraZeneca Gothenburg Pepparedsleden 1 SE-431 83 Mölndal Sweden Early Chemical Development Pharmaceutical Sciences R&D AstraZeneca Macclesfield SK10 2NA U.K. Data Science and Modelling Pharmaceutical Sciences R&D AstraZeneca Gothenburg Pepparedsleden 1 SE-431 83 Mölndal Sweden

来源：评论

学校读者我要写书评

暂无评论

Contextual and Nonstationary Multi-armed Bandits Using the Linear Gaussian State Space Model for the Meta-recommender System

Contextual and Nonstationary Multi-armed Bandits Using the L...

引用

2023 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2023

作者： Miyake, Yusuke Mine, Tsunenori Graduate School of ISEE Department of Advanced Information Technology Kyushu University Fukuoka Japan Pepabo R&D Institute GMO Pepabo Inc. Fukuoka Japan Faculty of Information Science and Electrical Engineering Fukuoka Japan

ISBN: (纸本)9798350337020

Selecting an optimal recommendation method is crucial for an electronic commerce (EC) site. However, the effectiveness of recommendation methods cannot be known in advance. Although continuous comparative evaluation in an actual environment is essential, it results in opportunity loss. To overcome this problem, opportunity loss reduction has been studied as a multi-armed bandit (MAB) problem, and adaptive meta-recommender systems (meta-rS) were devised to automatically and continuously select the best recommendation method according to a policy. The following three factors cause opportunity loss: the context of the recommendation method, temporal variation, and response time. However, studies are yet to formulate an MAB policy that considers all three factors. Thus, reducing opportunity loss remains a problem. We propose an MAB policy that considers all three causes of opportunity loss by using a Kalman filter for a linear Gaussian state space model. We conducted extensive experiments to select the best recommendation method using data from a real EC site. The results revealed that the proposed policy is highly effective for reducing opportunity loss of the meta-rS during evaluation and increasing cumulative clicks compared with baseline policies. © 2023 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Effects of drying time on the formation of merged and soft MAPbI3 grains and their photovoltaic responses

引用

Nanoscale Advances 2023年第8期5卷 2190-2198页

作者： Chandel, Anjali Ke, Qi Bin Chiang, Shou-En Cheng, Hsin-Ming Chang, Sheng Hsiung Department of Physics Chung Yuan Christian University Taiwan Taoyuan320314 Taiwan Research Center for Semiconductor Materials and Advanced Optics Taiwan Taoyuan320314 Taiwan Center for Nano Technology and R&D Center for Membrane Technology Chung Yuan Christian University Taiwan Taoyuan320314 Taiwan Department of Photonics National Cheng Kung University Taiwan Tainan701 Taiwan

The grain sizes of soft CH3NH3PbI3 (MAPbI3) thin films and the atomic contact strength at the MAPbI3/P3CT-Na interface are manipulated by varying the drying time of the saturated MAPbI3 precursor solutions, which influences the device performance and lifespan of the resultant inverted perovskite photovoltaic cells. The atomic-force microscopy images, cross-sectional scanning electron microscopy images, photoluminescence spectra and absorbance spectra show that the increased short-circuit current density (JSC) and increased fill factor (FF) are mainly due to the formation of merged MAPbI3 grains. Besides, the open-circuit voltage (VOC) of the encapsulated photovoltaic cells largely increases from 1.01 V to 1.15 V, thereby increasing the power conversion efficiency from 17.89% to 19.55% after 30 days, which can be explained as due to the increased carrier density of the MAPbI3 crystalline thin film. It is noted that the use of the optimized drying time during the spin coating process results in the formation of merged MAPbI3 grains while keeping the contact quality at the MAPbI3/P3CT-Na interface, which boosts the device performance and lifespan of the resultant perovskite photovoltaic cells. © 2023 rSC.

关键词： Photoluminescence

来源：评论

学校读者我要写书评

暂无评论

research on the Security Verification and Evaluation System of distribution Network dispatching Instructions in Telephone Mode 3

Research on the Security Verification and Evaluation System ...

引用

3rd International Conference on Energy, Power and Electrical Engineering, EPEE 2023

作者： Zhu, Chaoyue Huang, di Liu, Hongwei Chen, Panxia Luo, Shaojie Zheng, Weiyan Zhejiang Dayou Industrial Co. Ltd. Hangzhou Science and Technology Development Branch R&d Department Hangzhou China Business Service Center State Grid Hangzhou Power Supply Company Hangzhou China

ISBN: (纸本)9798350318180

Nowadays, the issue of command security in the distribution network dispatching telephone command mode is becoming increasingly prominent, and relevant personnel are paying more and more attention to the information security and accuracy when transmitting dispatching instructions to the distribution network, to prevent risks such as tampering, forgery, and theft of dispatching instruction information. The article first briefly analyzes the problems that exist in the telephone command mode during scheduling;Secondly, in response to the data security issues related to the issuance of instructions in the distribution network, based on cloud storage technology, the S-box is improved in the group encryption algorithm AES, and a chaotic system is introduced for data sequence encryption. The improved AES algorithm is combined with the chaotic system to achieve confidentiality of both parties' codes and information transmission;Simultaneously introducing upper entropy, using entropy to randomly verify the completeness and accuracy of numbers or information;Finally, the effectiveness of this method was theoretically proven, achieving efficient communication of scheduling information and ensuring the confidentiality and accuracy of received scheduling instructions. © 2023 IEEE.

关键词： cloud storage distribution network security verification and evaluation

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：