检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Xu, Yiming Shi, Bin Ma, Teng Dong, Bo Zhou, Haoyi Zheng, Qinghua Department of Computer Science and Technology Xi’an Jiaotong University China Shaanxi Provincial Key Laboratory of Big Data Knowledge Engineering Xi’an Jiaotong University China Department of Distance Education Xi’an Jiaotong University China School of Software Beihang University China Advanced Innovation Center for Big Data and Brain Computing Beihang University China

The graph with complex annotations is the most potent data type, whose constantly evolving motivates further exploration of the unsupervised dynamic graph representation. One of the representative paradigms is graph contrastive learning. It constructs self-supervised signals by maximizing the mutual information between the statistic graph’s augmentation views. However, the semantics and labels may change within the augmentation process, causing a significant performance drop in downstream tasks. This drawback becomes greatly magnified on dynamic graphs. To address this problem, we designed a simple yet effective framework named CLDG. Firstly, we elaborate that dynamic graphs have temporal translation invariance at different levels. Then, we proposed a sampling layer to extract the temporally-persistent signals. It will encourage the node to maintain consistent local and global representations, i.e., temporal translation invariance under the timespan views. The extensive experiments demonstrate the effectiveness and efficiency of the method on seven datasets by outperforming eight unsupervised state-of-the-art baselines and showing competitiveness against four semi-supervised methods. Compared with the existing dynamic graph method, the number of model parameters and training time is reduced by an average of 2,001.86 times and 130.31 times on seven datasets, respectively. The code and data are available at: https://***/yimingxu24/CLDG. © 2024, CC BY.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Integrated Gradients Demystified: An MRI Case Study on Aβ-T Protein Localization

Integrated Gradients Demystified: An MRI Case Study on Aβ-T...

引用

Metrology for Extended Reality, Artificial Intelligence and Neural engineering (MetroXRAINE), IEEE International Conference on

作者： Giorgio Dolci Cristian Morasso Federica Cruciani Lorenza Brusini Lorenzo Pini Vince D. Calhoun Ilaria Boscolo Galazzo Gloria Menegaz Dept. of Engineering for Innovation Medicine University of Verona Verona Italy Dept. of Computer Science University of Verona Verona Italy Padua Neuroscience Center University of Padua Padua Italy Tri-Institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS) Georgia State University Georgia Institute of Technology Emory University Atlanta GA USA

ISBN: (数字)9798350378009

ISBN: (纸本)9798350378016

Explainability in Artificial Intelligence is gaining increasing importance, especially in critical fields. In biomedical applications, attribution maps are particularly relevant for their inherent spatial localization property. Features that are relevant for a given task (e.g., classification of cases versus controls) can then be associated to specific regions, revealing disease-specific modulations and enabling ad-hoc post-hoc validation. Among these, baseline-dependent methods require a neutral image (called baseline) for computing the feature importance. The choice of the baseline is critical in that it determines the relevance map. In this work, we focused on a classification task, that is, subject stratification based on the amyloid-β and τ status in controls and patients in the Alzheimer's disease continuum, and employed Integrated Gradients to extract the attribution maps. This allowed identifying relevant brain regions driving the classification. Aware of the dependence of the outcomes from the chosen baseline, post-hoc statistical analysis was performed for validating the results confirming their plausibility. This puts forth the need to systematically assess the impact of the baseline for the correct interpretation of the corresponding outcomes.

关键词： Proteins Location awareness Protein engineering Statistical analysis Magnetic resonance imaging Neural engineering Modulation Metrology Feature extraction Reliability

来源：评论

学校读者我要写书评

暂无评论

A comparative study on shock compression and spall damage of three typical Cr–Ni stainless steels

引用

Materials Today Communications 2025年 46卷

作者： Lin, Z.H. Xu, J. Guo, S.H. Lu, L. Li, C. Liang, Xiangxiang Zhao, H.Y. Zhang, N.B. Cai, Y. Luo, S.N. Key Laboratory of Advanced Technologies of Materials Ministry of Education and Dynamic Materials Data Science Center Southwest Jiaotong University Chengdu Sichuan China The Peac Institute of Multiscale Sciences Chengdu Sichuan China Extreme Material Dynamics Technology Laboratory Chengdu Sichuan China School of Materials Science and Engineering Taiyuan University of Science and Technology Taiyuan Shanxi China Leshan Digital Intelligence and Innovation Institute Leshan Sichuan China

The shock Hugoniots, dynamic yield strength and spall damage of three typical Cr–Ni stainless steels (SS316L, SS304 and SS2025) are investigated via plate impact experiments. Free-surface velocity profiles are captured using laser Doppler velocimetry. The Hugoniots of single face-centered cubic (FCC) phase SS316L, SS304 with a predominant FCC phase, and dual-phase SS2205 are determined at shock stresses up to ∼18, 26 and 17 GPa, respectively. Among these three Cr–Ni stainless steels, SS2205 exhibits the highest dynamic yield stress, followed by SS316L and then SS304. Additionally, SS2205 demonstrates the highest spall strength, followed by SS304 and then SS316L. SS316L demonstrates a ductile spall damage behavior with a mixture of intergranular and transgranular damage, while ductile damage in SS304 is mainly governed by intergranular voids. In SS2205, both ductile voids and brittle cleavage cracks are associated with phase boundaries. Additionally, {111}〈112〉 deformation twins and newly formed martensitic laths are observed near the micro voids in SS316L and SS304, respectively. © 2025

关键词： Laser Doppler velocimeters

来源：评论

学校读者我要写书评

暂无评论

Mathematical Simulation of Linear Ubiquitination in T Cell Receptor-Mediated NF-κB Activation Pathway

Mathematical Simulation of Linear Ubiquitination in T Cell R...

引用

International Symposium on Fusion of Mathematics and Biology, 2020

作者： Oikawa, Daisuke Hatanaka, Naoya Suzuki, Takashi Tokunaga, Fuminori Department of Pathobiochemistry Graduate School of Medicine Osaka City University Osaka Japan Division of Mathematical Science Department of Systems Innovation Graduate School of Engineering Science Osaka University Toyonaka Japan Center for Mathematical Modeling and Data Science Osaka University Toyonaka Japan

ISBN: (纸本)9789811648656

The linear ubiquitin chain assembly complex (LUBAC), composed of the HOIP, HOIL-1L, and SHARPIN subunits, activates the canonical nuclear factor-κB (NF-κB) pathway through the Met1 (M1)-linked linear ubiquitination activity. On the course of the T cell receptor (TCR)-mediated NF-κB activation pathway, LUBAC transiently associates with and linearly ubiquitinates the CARMA1-BCL10-MALT1 (CBM) complex. In contrast, the linear ubiquitination of NEMO, a substrate of the TNF-α-induced NF-κB activation pathway, was limited in the TCR pathway. A linear ubiquitin-specific deubiquitinase (DUB), OTULIN, plays a major role in downregulating LUBAC-mediated TCR signaling. Mathematical modeling indicated that linear ubiquitination of the CBM complex accelerates the activation of IκB kinase (IKK), as compared with the activity induced by linear ubiquitination of NEMO alone. Moreover, simulations of the sequential linear ubiquitination of the CBM complex suggested that the allosteric regulation of linear (de)ubiquitination of CBM subunits is controlled by the ubiquitin-linkage lengths. Thus, unlike the TNF-α-induced NF-κB activation pathway, the TCR-mediated NF-κB activation in T cells has a characteristic mechanism to induce LUBAC-mediated NF-κB activation. © 2021, Springer Nature Singapore Pte Ltd.

关键词： Chemical activation

来源：评论

学校读者我要写书评

暂无评论

Performance analysis for a degrading system with Markov model 32nd

Performance analysis for a degrading system with Markov mode...

引用

32nd European Safety and Reliability Conference, ESREL 2022

作者： Zhang, Aibo Wu, Zhiying Wang, Yukun Xie, Min Center for Intelligent Multidimensional Data Analysis Hong Kong Science Park Shatin Hong Kong Department of Advanced Design and Systems Engineering City University of Hong Kong Hong Kong Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science and Innovation Chinese Academy of Sciences Hong Kong School of Economics and Management Tianjin Chengjian University Tianjin300384 China

ISBN: (纸本)9789811851834

Most mechanical engineering systems experience deterioration due to several practical factors such as aging, usage etc. This results in the degradation of its performance. Although the deterioration is continuous in nature, it is not always practically possible to measure the deterioration especially in the case of subsea systems. In such cases the performance of these systems is represented by multiple finite states in their lifetimes where various levels of degradation are represented by these states. Generally, the degradation process is divided into two stages, i.e., the normal and defective stages. Thus, there are three possible states for the system: normal, defective, and failed. The Markov approach has been utilized heavily for reliability analysis of such systems. However, most of the Markov models rely on an accurately exponentially distributed transition rate, which is unrealistic in certain cases. In reality, there are some uncertainties in the transition rates thanks to the limits of prior knowledge. This paper presents a multi-phase Markov model-based performance evaluation model for a three-state system with uncertainties in transition rates. The transition rates are assumed following Beta distribution with known lower and upper bounds. The average system unavailability in a given life cycle is employed to quantify system performance. © 2022 ESREL2022 Organizers. Published by Research Publishing, Singapore.

关键词： Reliability analysis

来源：评论

学校读者我要写书评

暂无评论

Enabling Intelligent Immersive Learning using Deep Learning-based Learner Confidence Estimation

Enabling Intelligent Immersive Learning using Deep Learning-...

引用

IEEE International Conference on Information Reuse and Integration (IRI)

作者： Mohammadreza Akbari Lor Shu-Ching Chen Mei-Ling Shyu Yudong Tao Shahin Vassigh School of Science and Engineering University of Missouri-Kansas City Kansas City MO USA Data Science and Analytics Innovation Center (dSAIC) University of Missouri-Kansas City Kansas City MO USA Department of Electrical and Computer Engineering University of Miami Coral Gables FL USA College of Communication Architecture and the Arts Florida International University Miami FL USA

ISBN: (数字)9798350351187

ISBN: (纸本)9798350351194

In today’s world, augmented reality and virtual reality (AR/VR) technologies have become more accessible to the public than ever. This brings the possibility of immersive learning to the forefront of education for future generations. However, there is still much to discover and improve in using these technologies to analyze and understand learning. This paper explores the utilization of data captured through AR/VR headsets during an immersive training program for industrial robotics. This includes data on time spent, eye gaze, and hand movement during a range of activities to track a learner’s understanding of the content and intelligently estimate learner confidence within these environments using deep learning. Leveraging a dataset that comprises responses and confidence levels from 10 individuals across 35 questions, we aim to improve the uses and applicability of confidence estimation. We explore the possibility of training a model using learners’ data to dynamically fine-tune lessons and activities for each individual, thereby improving performance. We demonstrate that a pre-trained compact LSTM classification model can be fine-tuned with relatively small data, for enhanced performance on an individual basis for better personalized learning.

关键词： Training Adaptation models Solid modeling Tracking Service robots Estimation data models Rough surfaces Long short term memory Immersive learning

来源：评论

学校读者我要写书评

暂无评论

Multimodal speech recognition using EEG and audio signals: A novel approach for enhancing ASR systems

引用

Smart Health 2024年 32卷

作者： Das, Anarghya Soni, Puru Huang, Ming-Chun Lin, Feng Xu, Wenyao Department of Computer Science and Engineering University at Buffalo Buffalo 14261 United States Department of Data and Computational Science Duke Kunshan University Jiangsu 215316 China ZJU-Hangzhou Global Scientific and Technological Innovation Center School of Cyber Science and Technology Zhejiang University Zhejiang 310007 China

Speech recognition using EEG signals captured during covert (imagined) speech has garnered substantial interest in Brain–Computer Interface (BCI) research. While the concept holds promise, current implementations must improve performance compared to established Automatic Speech Recognition (ASR) methods using audio. An area often underestimated in previous studies is the potential of EEG utilization during overt speech. Integrating overt EEG signals with speech data by leveraging advancements in deep learning presents significant potential to enhance the efficacy of these systems. This integration proves particularly advantageous in noisy environments and for individuals with speech impairments—challenges even conventional ASR techniques struggle to address effectively. Our investigation delves into this relationship by introducing a novel multimodal model that merges EEG and speech inputs. Our model achieves a multiclass classification accuracy of 95.39%. When subjected to artificial white noise added to the input audio, our model exhibits a notable level of resilience, surpassing the capabilities of models reliant solely on single EEG or audio modalities. The validation process, leveraging the robust techniques of t-SNE and silhouette coefficient, corroborates and solidifies these advancements. © 2024 Elsevier Inc.

关键词： Automatic Speech Recognition (ASR) Brain–Computer Interface (BCI) Deep learning Electroencephalography (EEG) Multimodal analysis Multimodal speech recognition Neural signal processing Overt speech

来源：评论

学校读者我要写书评

暂无评论

A New Kind of Singularly Degenerate Heteroclinic Cycle Revealed in a 3d Autonomous System

SSRN

引用

SSRN 2022年

作者： Su, Qifang Wang, Haijun Dong, Guili Pan, Jun Fan, Hongdan School of Electronics and Information Engineering School of Big Data Science Taizhou University Zhejiang Taizhou318000 China School of Innovation and Entrepreneurship Center of Engineering Training Zhejiang University of Science and Technology Hangzhou310023 China Department of Big Data Science School of Science Zhejiang University of Science and Technology Hangzhou310023 China

Being distinct from most known singularly degenerate heteroclinic cycles consisting of two different equilibria of a line or curve, or two parallel lines of semi-hyperbolic equilibria, a novel type of one that connects two perpendicular lines of semi-hyperbolic equilibria, i.e., $E_{z} = (0,0,z)$ and $E_{x} = (x,ex,k+ef)$, $z, x \in \mathbb{R}$, is found in the mathematical chaos model: $\dot{x} = a(y - x) + dxz$, $\dot{y} = kx + fy - xz$, $\dot{z} = -ex^{2} + xy + cz$ when $c=0$ and $e(a + fd) = (a - kd)$. Surprisingly, some three-scroll attractors are also created nearby that kind of singularly degenerate heteroclinic cycles in the case of small $c>0$.Further, some other rich dynamics is uncovered, i.e., global boundedness, Hopf bifurcation, limit cycles coexisting with one chaotic attractor, etc. We not only prove that the ultimate bound sets and globally exponentially attracting sets perfectly coincide under the same parameters, but also illustrate four limit cycles coexisting with one chaotic attractor, the saddle in the origin, and another two stable nontrivial node-foci, which are also trapped in the obtained globally exponentially attracting set. In addition, combining theoretical analysis and numerical simulation, the bidirectional forming mechanism of that kind of singularly degenerate heteroclinic cycle is illustrated, and the collapses of them may create three-scroll attractors, or limit cycles, etc. We expect that the study outcome may provide a reference on subsequent research. © 2022, The Authors. All rights reserved.

关键词： Lyapunov functions

来源：评论

学校读者我要写书评

暂无评论

OCT Medical Image Recognition Based on UNet++

OCT Medical Image Recognition Based on UNet++

引用

Chinese Control Conference (CCC)

作者： YunFei Bao Shangjie Ren Wei Zhang Hebei University of Technology Tianjin China Control Engineering Technology Innovation Center of Hebei Province Hebei University of Technology Tianjin P. R. China Tianjin University Tianjin China School of Artificial Intelligent and Data Science Control Engineering Technology Innovation Center of Hebei Province Hebei University of Technology Tianjin P. R. China

In this paper, we propose an improved UNet++ medical image segmentation method for coronary vascular OCT medical image detection to address the problem of large recognition errors arising from small targets and the limitation of arithmetic power in practical use when identifying stents and guidewire. Two layers of VGG feature extraction layer are used to extract the original image and input it into UNet++ model for cooperative training of multi-layer UNet model. Cross entropy loss and dice loss were used for evaluation, mixing the weight of the output of the multi-layer network model by a given strategy. The multi-objective semantic segmentation with high accuracy can be achieved while reducing the arithmetic power consumption of the model, Meanwhile, according to the requirements of different arithmetic power and recognition speed under the deployment of the model, the UNet model embedded in the model can be removed by pruning operation to reduce the arithmetic power consumption.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Detection of the Anaerobic Threshold from Cardiopulmonary Exercise Test data with EO-Attention-LSTM

Detection of the Anaerobic Threshold from Cardiopulmonary Ex...

引用

Chinese Control Conference (CCC)

作者： Xin Guo Zhaoxin Lun Qitong Chu School of Artificial Intelligence and Data Science Hebei University of Technology Tianjin China Qinhuangdao Research Institute National Research Center for Rehabilitation Technical Aids Qinhuangdao China Control Engineering Technology Innovation Center of Hebei Province Hebei University of Technology Tianjin China

The physiological indicators at the anaerobic threshold play an essential role in guiding scientific exercise. The gold standard methodology for detecting the anaerobic threshold is expert visual inspection, but its efficiency is low. Given the above problems, we propose an anaerobic threshold detection method combining the equilibrium optimizer (EO) algorithm and Attention-LSTM, which was verified on the data of 334 cardiopulmonary exercise tests. Through comparative experiments, the coefficient of determination (R2) of the Attention-LSTM anaerobic threshold detection model is 0.797, the R2 of the Attention-LSTM model after hyperparameter optimization by EO is 0.834. The results show that the detection error of the anaerobic threshold detection model proposed in this paper is consistent with that of experts, which can provide a certain reference value.

关键词：

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：