检索结果-内蒙古大学图书馆

SSRN 2025年

作者： Chen, Yinsong Yu, Samson S. Li, Zhong Eshraghian, Jason K. Lim, Chee Peng School of Engineering Deakin University MelbourneVIC3216 Australia Faculty of Mathmatics and Computer Science FernUniversität Hagen58084 Germany Department of Electrical and Computer Engineering University of California Santa CruzCA95064 United States Department of Computing Technologies Swinburne University of Technology HawthornVIC3122 Australia

While deep learning models have seen significant success across various domains, their black-box learning nature and lack of interpretability affect their reliability in safety-critical applications like medical diagnostics and autonomous vehicles. In an attempt to address these limitations, Bayesian neural networks (BNNs) offer a promising alternative by incorporating uncertainty estimation into model predictions, enhancing transparency and decision-making. However, BNN development has primarily focused on efficient, high-fidelity approximate inference and guaranteed convergence in asymptotic settings. These are unsuitable for modern high-dimensional, multimodal, and non-asymptotic deep learning applications, undermining their theoretical advantages. To bridge this gap, this paper provides in-depth reviews on how approximate Bayesian inference leverages deep learning optimization to achieve high efficiency and fidelity in high-dimensional spaces and multimodal loss landscapes. It also reconciles Bayesian consistency with generalization objectives in non-asymptotic settings and investigates the generalization capabilities of BNNs. Additionally, this survey examines the often-overlooked expressiveness of BNNs, emphasizing how weight uncertainty and the absence of in-between uncertainty affect their performance. This survey aims to inspire BNN practitioners to adopt a deep learning perspective and offer valuable insights to propel further advancements in the field. © 2025, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

GENERALIZED OXTOBY SUBSHIFTS AND HYPERFINITENESS

arXiv

引用

arXiv 2025年

作者： Deka, Konrad Peng, Bo Faculty of Mathematics and Computer Science Jagiellonian University in Krakow ul. Lojasiewicza L 6 Kraków30-348 Poland Department of Mathmatics and Statistics McGill University 805 Sherbrooke Street West MontrealQCH3A 2K6 Canada

We show that there exists a class of symbolic subshifts which realizes all Choquet simplices as simplices of invariant measures and the conjugacy relation on that class is *** Codes 03E15 © 2025, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Introducing QuantConn: Overcoming Challenging Diffusion Acquisitions with Harmonization 15th

Introducing QuantConn: Overcoming Challenging Diffusion Acqu...

引用

15th International Workshop on Computational Diffusion MRI, CDMRI 2024, held in conjunction with 27th International Conference on Medical Image Computing and computer-Assisted Intervention, MICCAI 2024

作者： Newlin, Nancy Schilling, Kurt Koudoro, Serge Chandio, Bramsh Qamar Kanakaraj, Praitayini Moyer, Daniel Kelly, Claire E. Genc, Sila Yang, Joseph Yuan-Mou Wu, Ye Adluru, Nagesh Nath, Vishwesh Pathak, Sudhir Schneider, Walter Gade, Anurag Consagra, William Rathi, Yogesh Hendriks, Tom Vilanova, Anna Chamberland, Maxime Pieciak, Tomasz Ciupek, Dominika Vega, Antonio Tristán Aja-Fernández, Santiago Malawski, Maciej Ouedraogo, Gani Machnio, Julia Thompson, Paul M. Jahanshad, Neda Garyfallidis, Eleftherios Landman, Bennett Department of Computer Science Vanderbilt University NashvilleTN United States Department of Radiology and Radiological Sciences Vanderbilt University Medical Center Nashville United States Indiana University Bloomington BloomingtonIN United States Mark and Mary Stevens Neuroimaging and Informatics Institute Keck School of Medicine of USC Los AngelesCA United States Developmental Imaging Murdoch Children’ Research Institute Melbourne Australia Murdoch Children’s Research Institute Melbourne Australia Turner Institute for Brain and Mental Health School of Psychological Sciences Monash University Melbourne Australia Department of Neurosurgery Royal Children’s Hospital Melbourne Australia Neuroscience Research Murdoch Children’s Research Institute Melbourne Australia Department of Paediatrics University of Melbourne Melbourne Australia School of Computer Science and Technology Nanjing University of Science and Technology Nanjing China Waisman Center Department of Radiology University of Wisconsin Madison United States NVIDIA Santa Clara United States Learning Research and Development Center University of Pittsburgh Pittsburgh United States Brigham and Women’s Hospital Boston United States Brigham and Women’s Hospital Harvard Medical School Boston United States Department of Computer Science and Mathmatics Eindhoven University of Technology Eindhoven Netherlands LPI ETSI Telecomunicación Universidad de Valladolid Castilla y León Spain Sano Centre for Computational Medicine Kraków30-054 Poland Aix-Marseille Université Marseille France Vanderbilt University Institute of Imaging Science Vanderbilt University NashvilleTN United States Department of Electrical and Computer Engineering Vanderbilt University NashvilleTN United States

ISBN: (纸本)9783031869198

White matter alterations are increasingly implicated in neurological diseases and their progression. Diffusion-weighted magnetic resonance imaging (DW-MRI) has been included in many international-scale studies to identify alterations in white matter microstructure and connectivity. Yet, quantitative investigation of DW-MRI data is hindered by a lack of consistency due to variations in acquisition protocols, sites, and scanners. Specifically, there is a need to harmonize the preprocessing of DW-MRI datasets to ensure that compatible and reproducible quantitative metrics are derived from each site, including (1) bundle-wise microstructure measures, (2) features of white matter fiber bundles, and (3) connectomics measures. In the MICCAI CDMRI 2023 QuantConn challenge, participants are provided raw data from the same individuals taken with two different acquisition protocols on a single 4 tesla scanner in the same scanning session and asked to preprocess the data in order to minimize acquisition differences while retaining biological variation. Here, we outline the testing framework, provide baseline pre-harmonized results, and discuss the learning implications of this challenge. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Photointerpretation

来源：评论

学校读者我要写书评

暂无评论

Digital Image Processing Using Noise Removal Technique: A Non-Linear Approach

Digital Image Processing Using Noise Removal Technique: A No...

引用

2022 IEEE Asia-Pacific Conference on computer science and Data Engineering, CSDE 2022

作者： Chand, Rishal Ravikesh Farik, Mohammed Sharma, Neeraj Anand School of Science and Technology The University of Fiji Department of Computer Science and Mathmatics Fiji

ISBN: (纸本)9781665453059

An undesirable by-product of a captured image is Image Noise-random distribution of brightness and color misinformation. Denoising is one of the common components which is present in digital image processing techniques, Image processing has become an integral part of today's era amongst rapidly growing technologies. The process of manipulating images by using various algorithms exploring techniques to pre-process an image, improving its pictorial information for human interpretation. This research paper explores different types of image noise that can be present in a digital image and degradation by photo electronic noise, impulse (salt & pepper) noise and gaussian noise, thus reviews on linear approaches to denoise the image using 2D convolution filters and median non-linear filters. © 2022 IEEE.

关键词： Salt and pepper noise

来源：评论

学校读者我要写书评

暂无评论

New Operator for Image Edge Detection Based on Finite Difference Method 6

New Operator for Image Edge Detection Based on Finite Differ...

引用

6th International Conference on Intelligent Computing in Data sciences, ICDS 2024

作者： Ghosn, Ali El-Zaart, Ali El-Arwadi, Toufic Beirut Arab University Faculty of Science Department of Mathmatics and Computer Science Beirut Lebanon

ISBN: (纸本)9798350351200

Detection of edges in an image is a very important step towards understanding the image features and it is considered as the most common approach for detecting meaningful discontinuities in gray level. The edges in an image are detected using the derivative of the image. The method that uses the first derivative is called gradient while the one that uses the second derivative is called Laplacian. This paper is concerned in the edge detection problem using the first derivative method. Our proposed method of edge detection is based on the implicit finite difference, where we have only used the first derivative of the images for x and y directions in order to construct the two masks. The proposed masks are more general than the classical gradient masks like Sobel, Prewitt, and Robert. Our proposed method has been implemented and tested on artificial images, and then compared with Sobel and Prewitt edge detectors using a performance measure. The results showed good detection of edges in low contrast images in comparison with Sobel and Prewitt edge detectors. © 2024 IEEE.

关键词： Image understanding

来源：评论

学校读者我要写书评

暂无评论

From descriptive to numeric emotions in agents

From descriptive to numeric emotions in agents

引用

2010 International Conference on Artificial Intelligence, ICAI 2010

作者： Jiang, Hong Mathmatics and Computer Science Department Benedict College Columbia SC United States

ISBN: (纸本)9781601321480

Representing emotions in agents is important for designing agents that behave like humans or are intended to interact with humans. OCC model, a descriptive emotion model, has been used widely as a standard for emotion synthesis, however it is also criticized for falling short on suggestions for what to do with the emotional states. PAD emotion scale, a numeric emotion measurement of emotions, supplies a highly useful and convenient assessment of consumer emotional reactions to services, products, or combinations of products and services. It closely connects human's emotional states with the behaviors. If we can find out the relationship between these two models, we can produce a more complete emotion mechanism by using OCC to generate emotional states and PAD to analyze them. Meanwhile, when we consider emotional negotiation, a numeric measurement of emotions is desirable. We need to find a proper way to embed emotions into a utility function. This paper presents a method to map descriptive OCC emotions to numeric PAD emotions, thereby providing an opportunity to embed both the OCC model and the PAD emotion scale into emotional agents to produce a more complete emotion mechanism. The mechanism includes both generating emotions and predicting the effects of emotions on agents' reasoning processes and behaviors. More specifically, we use WordNet as a lexical ontology and apply ontology matching techniques to produce a mapping from the 22 basic descriptive emotions generated by the OCC model to the numeric PAD emotion space. The result is consistent with the senses in WordNet and the situations in the OCC model.

关键词： Ontology

来源：评论

学校读者我要写书评

暂无评论

Smoke 100k: A database for smoke detection 8

Smoke 100k: A database for smoke detection

引用

8th IEEE Global Conference on Consumer Electronics, GCCE 2019

作者： Cheng, Hsiang-Yin Yin, Jia-Li Chen, Bo-Hao Yu, Zhi-Min Yuan Ze University Department of Computer Science and Engineering Taoyuan320 Taiwan College of Mathmatics and Computer Science Fuzhou University Fuzhou350116 China

ISBN: (纸本)9781728135755

Due to the complex scenarios and the limited feature information in a single image, a precise smoke detection is much more challenging in practice. Most of previous smoke detection methods either extract textural and spatiotemporal characteristics of smoke or separate the smoke and background components of the image. However, those methods often fail in detecting smoke positions because of the limited feature information within a single image. Moreover, the task of smoke detection can be better achieved if the extra information from collected training dataset is available. One key issue is how to build a training dataset of paired smoke images and ground-truth bounding box positions for end-to-end learning. This paper proposes a large-scale benchmark image dataset to train a smoke detector. With the built dataset, experimental results demonstrate that the discriminative models can be effectively trained as the smoke detector to detect the smoldering fires precisely. © 2019 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

The Application of Coupled Three Order Cumulants’ Differential Feature in Fault Diagnosis

引用

International Journal of Modern Nonlinear Theory and Application 2018年第2期7卷 77-87页

作者： Wenbing Wu Jinquan Xiong Yupeng Wu Rihua Liu Department of Mathmatics and Computer Science Nanchang Normal University Nanchang China

Complex three-order cumulant has different definition forms. Different forms conclude different information. For studying the effection of frequency in the coupled signals to fault diagnosis, the differential method to the three order cumulants of coupled signals is adopted. By using the differential of complex three order cumulants before and after respectively, then their ?dimensional spectrum is calculated, and the results are used to fault diagnosis. The experimental results show that, the increase frequency item in three order cumulants after differentiated impacts on the results of fault diagnosis and the degree of effection is relative to the differential times. And the correct rate of fault diagnosis can be raised by changing the differential times of three order cumulants.

关键词： Three Order Cumulants Differential Coupled 1½ Dimensional Spectrum Fault Diagnosis

来源：评论

学校读者我要写书评

暂无评论

The Application of Cumulants Slices in Fault Diagnosis

引用

International Journal of Modern Nonlinear Theory and Application 2018年第3期7卷 89-96页

作者： Wenbing Wu Department of Mathmatics and Computer Science Nanchang Normal University Nanchang China

Complex third-order cumulant has different definition forms. Different forms have different coupling properties, and the generated complex cumulants slices contain different coupling information of signals. In experiments, using the different definitions, the same coupling method is applied to both specific fault signals and normal signals. Furthermore, complex third-order cumulant slices spectrum is defined, and it is used to analyse the coupling features of normal signals and fault signals. Experiments indicate that the detection accuracy rate on the same fault is not the same with the different coupling method, thus, it provides an alternative method to diagnose the specific fault.

关键词： Third-Order Cumulant Slices Fault Diagnosis Spectrum of Slices

来源：评论

学校读者我要写书评

暂无评论

Performance of Data Images Encryption Based Karhunen-Leoeve Transforms

引用

Research Journal of Applied sciences, Engineering and Technology 2017年第3期14卷 95-100页

作者： Hind Rustum Mohammed Shaban Inaam R. Alsaiq Alaa Abdul Hussein Mezher Computer Science Department Mathmatics Department Faculty of Computer Science and Maths Kufa University Najaf 54001 Iraq

The purpose of this study encryption method will apply on all parts image by the Karhunen-Loeve Transform (KLT). It will be discussed time complexity and storage complexity in method (database consists of 40 images). The Karhunen-Loeve Transform (KLT) is always used in image processing and it has a wide application area. It was called ＂the best transform＂ because the decorrelation, energy concentration and Mean Square Error (MSE) is very small. The first step is found The Characteristics of the (KLT) by convert image matrix to elements vector, the second step found the covariance matrix, then compute its Great ability of decorrelation. Finally, we will reconstruct the de-nosing matrix to the Image by KLT. The experimental results and security analysis confirm the effectiveness of the indicates the robustness and advantages of the algorithm.

关键词： Encryption algorithms encryption data images encryption decryption algorithm karhunen-leoeve transforms mathematics module

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：