检索结果-内蒙古大学图书馆

IEEE International Symposium on Information Theory

作者： Panyu Zhu Jingjie Lv Yunghsiang S. Han Linqi Song Hanxu Hou Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China School of Electrical Engineering & Intelligentization Dongguan University of Technology Department of Computer Science City University of Hong Kong

ISBN: (数字)9798350382846

ISBN: (纸本)9798350382853

EVENODD+ codes are binary maximum distance separable (MDS) array codes for correcting double disk failures in RAID-6 with asymptotically optimal encoding/decoding/update complexities. However, the number of bits stored in each disk of EVENODD+ codes should be an odd number minus one. In this paper, we present a new construction of EVENODD+ codes that have more flexible parameters. The number of bits stored in each disk of our codes is an odd minus one times any positive integer. Moreover, our codes not only have asymptotically optimal encoding/decoding/update complexities but also have lower encoding/decoding/update complexities than the existing EVENODD+ codes.

关键词： Codes Complexity theory

来源：评论

学校读者我要写书评

暂无评论

nmODE-MVC: Neural Memory ODE-Based Multi-View Clustering

nmODE-MVC: Neural Memory ODE-Based Multi-View Clustering

引用

Artificial Intelligence technology (ACAIT), Asian Conference on

作者： Yan Zheng Song Wu Yazhou Ren Xiaorong Pu Zenglin Xu Lifang He Zhang Yi School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China School of Computer Science and Engineering Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China School of Computer Science and Technology Harbin Institute of Technology Shenzhen Shenzhen China Department of Computer Science and Engineering Lehigh University Bethlehem USA Intelligent Interdisciplinary Research Center and College of Computer Science Sichuan University Chengdu China

ISBN: (数字)9798350359145

ISBN: (纸本)9798350359152

Deep multi-view clustering (MVC) has recently gained significant interest for its capability to harness complementary information across multiple views through deep neural networks, enhancing clustering performance. However, existing deep MVC methods still have two issues: (1) They separate the feature learning objective and the clustering objective, potentially leading to suboptimal embedding features due to the neglect of the global clustering structure. (2) Most works utilize the target distribution obtained by the clustering objective to supervise feature learning. However, they simply sharpen the soft clustering distribution to obtain the target distribution, lacking the dynamics of the clustering process. To address these issues, in this paper we propose a novel end-to-end self-supervised multi-view clustering approach based on NeuralODEs, namely nmODE-MVC. Our approach operates in two stages: In the first stage, all view-embedded features are extracted, and then unified them into a global feature, facilitating global clustering assignments by exploiting multi-view data’s complementarity. The second stage leverages the dynamic attributes of NeuralODEs to dynamically refine the data features and generate a target distribution that actively drives the clustering decisions. We formulate these learning processes as a unified optimization problem, facilitating iterative training and refinement. The effectiveness of nmODE-MVC is validated through extensive experiments on multiple real-world multi-view datasets.

关键词： Representation learning Training Learning (artificial intelligence) Artificial neural networks Drives Feature extraction Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Efficient Data-aware Distance Comparison Operations for High-Dimensional Approximate Nearest Neighbor Search

arXiv

引用

arXiv 2024年

作者： Deng, Liwei Chen, Penghao Zeng, Ximu Wang, Tianfu Zhao, Yan Zheng, Kai University of Electronic Science and Technology of China China University of Science and Technology of China China School of Computer Science and Engineering University of Electronic Science and Technology of China China Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China China

High-dimensional approximate k nearest neighbor search (AKNN) is a fundamental task for various applications, including information retrieval. Most existing algorithms for AKNN can be decomposed into two main components, i.e., candidate generation and distance comparison operations (DCOs). While different methods have unique ways of generating candidates, they all share the same DCO process. In this study, we focus on accelerating the process of DCOs that dominates the time cost in most existing AKNN algorithms. To achieve this, we propose an Data-Aware Distance Estimation approach, called DADE, which approximates the exact distance in a lower-dimensional space. We theoretically prove that the distance estimation in DADE is unbiased in terms of data distribution. Furthermore, we propose an optimized estimation based on the unbiased distance estimation formulation. In addition, we propose a hypothesis testing approach to adaptively determine the number of dimensions needed to estimate the exact distance with sufficient confidence. We integrate DADE into widely-used AKNN search algorithms, e.g., IVF and HNSW, and conduct extensive experiments to demonstrate the superiority. Copyright © 2024, The Authors. All rights reserved.

关键词： Nearest neighbor search

来源：评论

学校读者我要写书评

暂无评论

An Early-Termination Method for the Welch-Berlekamp Algorithm

An Early-Termination Method for the Welch-Berlekamp Algorith...

引用

2023 IEEE International Symposium on Information Theory, ISIT 2023

作者： Chen, Chao Han, Yunghsiang S. Tang, Nianqi Lin, Sian-Jheng Bai, Baoming Ma, Xiao Xidian University State Key Laboratory of Isn Xi'an China University of Electronic Science and Technology of China Shenzhen Institute for Advanced Study China Huawei Technologies Co. Ltd. China Sun Yat-sen University School of Computer Science and and Engineering Guangzhou China

ISBN: (纸本)9781665475549

This paper presents an early-termination method for the Welch-Berlekamp algorithm. Specifically, if e = t (where e is the number of errors and t is the error correction capability), the Welch-Berlekamp algorithm can be terminated at latest at the completion of the (t + e)-th iteration. Based on the early-termination mechanism, a new variant of the Welch-Berlekamp algorithm called eFDMA is presented, and a systolic architecture is designed for the eFDMA algorithm. This provides an efficient implementation for the key equation solver for a new class of Reed-Solomon codes recently proposed by Lin et al. [9]. © 2023 IEEE.

关键词： Reed-Solomon codes

来源：评论

学校读者我要写书评

暂无评论

Fast Encoding of Extended Integrated Interleaved Codes Based on Two-Dimensional LCH-FFT 12

Fast Encoding of Extended Integrated Interleaved Codes Based...

引用

12th International Symposium on Topics in Coding, ISTC 2023

作者： Chen, Chao Caiy, Suihua Han, Yunghsiang S. Tang, Nianqi Bai, Baoming Xidian University State Key Laboratory of Isn Xi'an China Sun Yat-sen University School of Computer Science and and Engineering Guangzhou China University of Electronic Science and Technology of China Shenzhen Institute for Advanced Study China Huawei Technologies Co. Ltd. China

ISBN: (纸本)9798350326116

In this paper, we present fast encoding algorithms for extended integrated interleaved (EII) codes by re-defining them as a class of bivariate polynomial evaluation codes. Based on the recently proposed two-dimensional LCH-FFT algorithm, fast algorithms are derived for nonsystematic and systematic encoding. For an EII code of length N, both the encoding algorithms achieve the computational complexity O(N log (N)). © 2023 IEEE.

关键词： Fast Fourier transforms

来源：评论

学校读者我要写书评

暂无评论

Fast Erasure Decoding of Reed-Solomon Codes Based on Decomposition of Vandermonde Matrix

Fast Erasure Decoding of Reed-Solomon Codes Based on Decompo...

引用

International Conference on Wireless Communications and Signal Processing (WCSP)

作者： Zebing Lin Jingjie Lv Pingping Li Linqi Song Hui Liang Hanxu Hou School of Computer Science and Technology Dongguan University of Technology School of Electrical Engineering & Intelligentization Dongguan University of Technology Shenzhen Institute for Advanced Study University of Electronic Science & Technology of China Department of Computer Science City University of Hong Kong City University of Hong Kong Shenzhen Research Institute

ISBN: (数字)9798350390643

ISBN: (纸本)9798350390650

Reed-Solomon (RS) codes are widely applied in storage systems. The main bottleneck of RS codes is that the high decoding complexity defined as the number of multiplications and additions. In this paper, we propose a fast decoding algorithm for RS codes that requires fewer number of multiplications. Our fast decoding algorithm is designed by a decomposition of Vandermonde matrices. We show that our decoding algorithm requires fewer number of multiplications than the existing best known algorithm when the number of parities $T$ is large. When the codeword length $N=16$ , our decoding algorithm reduces the number of multiplications by 7% and 29% when $T=4$ and $T=7$ , respectively, compared with the best known decoding algorithm.

关键词： Reed-Solomon codes Wireless communication Codes Upper bound Signal processing algorithms Transforms Signal processing Decoding Matrix decomposition Galois fields

来源：评论

学校读者我要写书评

暂无评论

Human Expansion-Induced Biodiversity Crisis over Asia from 2000 to 2020

引用

Research 2024年第2期2023卷 647-658页

作者： Chao Yang Qingquan Li Xuqing Wang Aihong Cui Junyi Chen Huizeng Liu Wei Ma Xuanyan Dong Tiezhu Shi Fanyi Meng Xiaohu Yan Kai Ding Guofeng Wu MNR Key Laboratory for Geo-Environmental Monitoring of Great Bay Area&Guangdong Key Laboratory of Urban Informatics&Shenzhen Key Laboratory of Spatial Smart Sensing and Services Shenzhen UniversityShenzhen 518060China School of Architecture and Urban Planning Shenzhen UniversityShenzhen 518060China College of Civil and Transportation Engineering Shenzhen UniversityShenzhen 518060China Guangdong Laboratory of Artificial Intelligence and Digital Economy(SZ) Shenzhen 518107China Center for Hydrogeology and Environmental Geology China Geological SurveyBaoding 071051China Department of Geography Hong Kong Baptist UniversityHong Kong Special Administrative Region 999077China Faculty of Land Resource Engineering Kunming University of Science and TechnologyKunming 650093China institute for Advanced Study and Tiandu-Shenzhen University Deep Space Joint Laboratory Shenzhen UniversityShenzhen 518060China School of Civil Engineering Chongqing Jjiaotong UniversityChongging 400074China Department of Cil and Environmental Engineering Tohoku UniversitySendai 980-8579Japan School of Artificial Intelligence Shenzhen PolytechnicShenzhen 518055China School of Computer Science and Technology Dongguan University of TechnologyDongguan 523419China

Asia stands out as a priority for urgent biodiversity conservation due to its large protected areas(PAs)and threatened *** the 21st century,both the highlands and lowlands of Asia have been experiencing the dramatic human ***,the threat degree of human expansion to biodiversity is poorly ***,the threat degree of human expansion to biodiversity over 2000 to 2020 in Asia at the continental(Asia),national(48 Asian countries),and hotspot(6,502 Asian terrestrial PAs established before 2000)scales is investigated by integrating multiple large-scale *** results show that human expansion poses widespread threat to biodiversity in Asia,especially in Southeast Asia,with Malaysia,Cambodia,and Vietnam having the largest threat degrees(~1.5 to 1.7 times of the Asian average level).Human expansion in highlands induces higher threats to biodiversity than that in lowlands in one-third Asian countries(most Southeast Asian countries).The regions with threats to biodiversity are present in~75%terrestrial PAs(including 4,866 PAs in 26 countries),and human expansion in PAs triggers higher threat degrees to biodiversity than that in *** findings provide novel insight for the Sustainable Development Goal 15(SDG-15 Life on Land)and suggest that human expansion in Southeast Asian countries and PAs might hinder the realization of *** reduce the threat degree,Asian developing countries should accelerate economic transformation,and the developed countries in the world should reduce the demands for commodity trade in Southeast Asian countries(i.e.,trade leading to the loss of wildlife habitats)to alleviate human expansion,especially in PAs and highlands.

关键词： Human terrestrial integrating

来源：评论

学校读者我要写书评

暂无评论

Domain Adaptive Detection of MAVs: A Benchmark and Noise Suppression Network

arXiv

引用

arXiv 2024年

作者： Zhang, Yin Deng, Jinhong Liu, Peidong Li, Wen Zhao, Shiyu The College of Computer Science and Technology Zhejiang University Hangzhou310027 China The School of Engineering Westlake University Hangzhou310024 China The School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu611731 China The Research Center for Industries of the Future School of Engineering Westlake University Westlake Institute for Advanced Study Hangzhou310024 China

Visual detection of Micro Air Vehicles (MAVs) has attracted increasing attention in recent years due to its important application in various tasks. The existing methods for MAV detection assume that the training set and testing set have the same distribution. As a result, when deployed in new domains, the detectors would have a significant performance degradation due to domain discrepancy. In this paper, we study the problem of cross-domain MAV detection. The contributions of this paper are threefold. 1) We propose a Multi-MAVMulti-Domain (M3D) dataset consisting of both simulation and realistic images. Compared to other existing datasets, the proposed one is more comprehensive in the sense that it covers rich scenes, diverse MAV types, and various viewing angles. A new benchmark for cross-domain MAV detection is proposed based on the proposed dataset. 2) We propose a Noise Suppression Network (NSN) based on the framework of pseudo-labeling and a large-to-small training procedure. To reduce the challenging pseudo-label noises, two novel modules are designed in this network. The first is a prior-based curriculum learning module for allocating adaptive thresholds for pseudo labels with different difficulties. The second is a masked copy-paste augmentation module for pasting truly-labeled MAVs on unlabeled target images and thus decreasing pseudo-label noises. 3) Extensive experimental results verify the superior performance of the proposed method compared to the state-of-the-art ones. In particular, it achieves mAP of 46.9%(+5.8%), 50.5%(+3.7%), and 61.5%(+11.3%) on the tasks of simulation-to-real adaptation, cross-scene adaptation, and cross-camera adaptation, respectively. Note to Practitioners-To study the cross-domain MAV detection problem, this paper establishes a novel benchmark that consists of three domain adaptation tasks: simulation-to-real adaptation, cross-scene adaptation, and cross-camera adaptation, respectively. The benchmark is based on a novel MAV da

关键词： Spurious signal noise

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Adaptive Lesion Segmentation with Similarity Based Cross Pseudo Supervision

Unsupervised Adaptive Lesion Segmentation with Similarity Ba...

引用

Artificial Intelligence technology (ACAIT), Asian Conference on

作者： Yuxin Zhu Shuhao Li Aodi Yang Mei Feng Xiaorong Pu Yazhou Ren School of Computer Science and Engineering University of Electronic Science and Technology of China Department of Medical Oncology The Third People’s Hospital of Sichuan China School of Computer Science and Engineering Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China

ISBN: (数字)9798350359145

ISBN: (纸本)9798350359152

Notable advancement has been made in lesion segmentation based on deep learning. Nonetheless, the effectiveness of deep learning heavily relies on the availability of a substantial amount of high-quality labeled data, which is labor-intensive and expensive to require. Moreover, when applied to a different domain, deep learning-based lesion segmentation methods often experience a decline in performance due to variations in device vendors and patient populations. The unsupervised domain adaptation method is proposed to tackle the domain discrepancy problem. There are certain similarities between semi-supervised learning and unsupervised domain adaptation. From an application perspective, semi-supervised learning models can be used to solve unsupervised domain adaptation *** this end, we introduce the cross-pseudo-supervision mechanism in the semi-supervised field for unsupervised adaptive lesion segmentation. However, the method ignores the similar characteristics in medical images. To fully leverage the semantic similarity between domains, a novel similarity-based cross-pseudo-supervised lesion segmentation model is proposed. The proposed model employs a U-Net encoder, trained on source domain, to extract features from both the source and target domains. Subsequently, cosine similarity is computed based on these extracted features. The obtained similarity scores are then employed to weigh the source samples during the training process. The effectiveness of the proposed method is demonstrated through cross-domain experiments conducted on two publicly available COVID-19 lesion segmentation datasets.

关键词： Deep learning Training Adaptation models Image segmentation Computational modeling Semisupervised learning Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Analyzing the Effectiveness of Smartphone Applications on Disease Management and Healthcare Systems

Analyzing the Effectiveness of Smartphone Applications on Di...

引用

International Conference on Application of Information and Communication Technologies (AICT)

作者： Rabei Raad Ali Wisam Subhi Al-Dayyeni Christy Atika Sari Eko Hari Rachmawanto Candra Irawan Omar M. Jihad Department of Computer Engineering Technology Northern Technical University Mosul Iraq School of Information Technologies and Engineering ADA University Baku Azerbaijan Study Program in Informatics Engineering Faculty of Computer Science University of Dian Nuswantoro Semarang Indonesia Study Program in Information Systems Faculty of Computer Science University of Dian Nuswantoro Semarang Indonesia

ISBN: (数字)9798350387537

ISBN: (纸本)9798350387544

In today’s complex global environments, it is highly challenging to achieve effective healthcare management. The critical challenges faced by healthcare management are being met by the revolution of m-health where mobile cloud computing plays a major role. The taxonomy of mobile computing comprises operational aspects, end-user issues, and service quality and mobility management. The use of smartphone technologies and applications has become a highly significant approach to improving healthcare management. Mobile health services such as mobile pathology, mobile neurosurgery, cancer treatment, and behavioral/psychological disorders are gaining significance where smartphone applications are being used. Portability, flexibility, and convenience are major characteristics of mobile computing that have helped patients and doctors to develop better relationships through coordination and communication. The relationship between smartphone technologies and applications and healthcare management can be understood in several broad aspects. This article aims to analyze the current and future implications of these technologies on healthcare and disease management systems.

关键词： Pathology Cloud computing Taxonomy Cancer treatment Medical services Benchmark testing Neurosurgery Information and communication technology Mobile computing Diseases

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：