检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Gao, Chenqiang Liu, Chuandong Shu, Jun Liu, Fangcen Liu, Jiang Yang, Luyu Gao, Xinbo Meng, Deyu The School of Intelligent Systems Engineering The Shen-zhen Campus of Sun Yatsen University Sun Yat-sen University Guangdong Shenzhen518107 China The School of Computer Science Wuhan University Wuhan430072 China School of Mathematics and Statistics Ministry of Education Key Lab of Intelligent Networks and Network Security Xi’an Jiaotong University Shaanxi 710049 China School of Communication and Information Engineering Chongqing University of Posts and Telecommunications Chongqing400065 China Meta Menlo Park94025 United States University of Maryland College Park20742 United States

Current state-of-the-art (SOTA) 3D object detection methods often require a large amount of 3D bounding box annotations for training. However, collecting such large-scale densely-supervised datasets is notoriously costly. To reduce the cumbersome data annotation process, we propose a novel sparsely-annotated framework, in which we just annotate one 3D object per scene. Such a sparse annotation strategy could significantly reduce the heavy annotation burden, while inexact and incomplete sparse supervision may severely deteriorate the detection performance. To address this issue, we develop the SS3D++ method that alternatively improves 3D detector training and confident fully-annotated scene generation in a unified learning scheme. Using sparse annotations as seeds, we progressively generate confident fully-annotated scenes based on designing a missing-annotated instance mining module and reliable background mining module. Our proposed method produces competitive results when compared with SOTA weakly-supervised methods using the same or even more annotation costs. Besides, compared with SOTA fully-supervised methods, we achieve on-par or even better performance on the KITTI dataset with about 5× less annotation cost, and 90% of their performance on the Waymo dataset with about 15× less annotation cost. The additional unlabeled training scenes could further boost the performance. The code will be available at https://***/gaocq/SS3D2. Copyright © 2024, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Solution of secure multi-party multi-data ranking problem based on El Gamal encryption

引用

Tongxin Xuebao/Journal on Communications 2007年第11期28卷 1-5页

作者： Liu, Wen Luo, Shou-Shan Chen, Ping School of Computer Science and Technology Beijing University of Posts and Telecommunications Beijing 100876 China School of Software Engineering Beijing University of Posts and Telecommunications Beijing 100876 China National Key Lab. of Integrated Service Networks Xidian University Xi'an 710071 China School of Telecommunication Engineering Beijing University of Posts and Telecommunications Beijing 100876 China

Based on El Gamal homomorphic encryption, a protocol of secure multi-party multi-data ranking problem was proposed. The problem extended millionaires' problem. Furthermore, the correctness and security of this protocol were shown in semi-honest model, using the definition of secure multi-party computation. The solution of secure multi-party multi-data ranking problem can form a fundamental basis of new-style electronic transaction, such as private biding and auction, secret online transaction and so on.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

Retraction Note: Cybersecurity enhancement to detect credit card frauds in health care using new machine learning strategies

引用

Soft Computing 2024年第1期28卷 331-331页

作者： Jayanthi, E. Ramesh, T. Kharat, Reena S. Veeramanickam, M. R. M. Bharathiraja, N. Venkatesan, R. Marappan, Raja Networks and IoT Lab Department of Computer Science and Engineering Presidency University Bangalore India Department of Computer Science & Engineering R.M.K Engineering College Chennai India Pimpri Chinchwad College of Engineering Pune India Chitkara University Institute of Engineering and Technology Chitkara University Rajpura India SASTRA Deemed University Thanjavur India

来源：评论

学校读者我要写书评

暂无评论

Demonstration of a Task-flow based Aircraft Collaborative Design Application in Optical Grid

Demonstration of a Task-flow based Aircraft Collaborative De...

引用

33rd European Conference and Exhibition of Optical Communication

作者： Zhengyu Wang Wei Guo Zhenyu Sun Yaohui Jin Weiqiang Sun Weisheng Hu Xinhua Lin Min-You Wu Hong Liu San Fu Jun Yuan Chunming Qiao State Key Lab on Fiber-Optic Local Area Networks and Advanced Optical Communication systems Shanghai Jiao Tong University P. R. China 200240 Grid Computing Center Shanghai Jiao Tong University China 200240 Institute of Aerospace Science and Technology Shanghai Jiao Tong University China 200240 Shanghai Supercomputer Center Shanghai China 201203 Computer Science & Engineering State University of New York at Buffalo NY 14260 USA

One task-flow scheduler has been designed to optimally allocate resources for a task-flow based Aircraft Collaborative Design Application in Optical Grid. This scheduler and application has been deployed on an Optical... 详细信息

关键词： Computational modeling Atmospheric modeling Numerical models Performance evaluation Optical design Aerodynamics

来源：评论

学校读者我要写书评

暂无评论

Disentangled Noisy Correspondence Learning

arXiv

引用

arXiv 2024年

作者： Dang, Zhuohang Luo, Minnan Wang, Jihong Jia, Chengyou Han, Haochen Wan, Herun Dai, Guang Chang, Xiaojun Wang, Jingdong The School of Computer Science and Technology The Ministry of Education Key Laboratory of Intelligent Networks and Network Security The Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi'an Jiaotong University Shaanxi Xi'An710049 China SGIT AI Lab State Grid Shaanxi Electric Power Company Limited State Grid Corporation of China Shaanxi China The School of Information Science and Technology University of Science and Technology China United Arab Emirates The Baidu Inc China

Cross-modal retrieval is crucial in understanding latent correspondences across modalities. However, existing methods implicitly assume well-matched training data, which is impractical as real-world data inevitably involves imperfect alignments, i.e., noisy correspondences. Although some works explore similarity-based strategies to address such noise, they suffer from sub-optimal similarity predictions influenced by modality-exclusive information (MEI), e.g., background noise in images and abstract definitions in texts. This issue arises as MEI is not shared across modalities, thus aligning it in training can markedly mislead similarity predictions. Moreover, although intuitive, directly applying previous cross-modal disentanglement methods suffers from limited noise tolerance and disentanglement efficacy. Inspired by the robustness of information bottlenecks against noise, we introduce DisNCL, a novel information-theoretic framework for feature Disentanglement in Noisy Correspondence Learning, to adaptively balance the extraction of MII and MEI with certifiable optimal cross-modal disentanglement efficacy. DisNCL then enhances similarity predictions in modality-invariant subspace, thereby greatly boosting similarity-based alleviation strategy for noisy correspondences. Furthermore, DisNCL introduces soft matching targets to model noisy many-to-many relationships inherent in multi-modal input for noise-robust and accurate cross-modal alignment. Extensive experiments confirm DisNCL's efficacy by 2% average recall improvement. Mutual information estimation and visualization results show that DisNCL learns meaningful MII/MEI subspaces, validating our theoretical analyses. Copyright © 2024, The Authors. All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Investigating Bi-level optimization for learning and vision from a unified perspective: A survey and beyond

arXiv

引用

arXiv 2021年

作者： Liu, Risheng Gao, Jiaxin Zhang, Jin Meng, Deyu Lin, Zhouchen DUT-RU International School of Information Science & Engineering Dalian University of Technology Key Laboratory for Ubiquitous Network Service Software of Liaoning Province Dalian116024 China Department of Mathematics Southern University of Science and Technology National Center for Applied Mathematics Shenzhen China School of Mathematics and Statistics Ministry of Education Key Lab of Intelligent Networks and Network Security Xi’an Jiaotong University Xi’an Shaanxi China School of Electronics Engineering and Computer Science Peking University Beijing100871 China Cooperative Medianet Innovation Center Shanghai Jiao Tong University Shanghai200240 China

Bi-Level Optimization (BLO) is originated from the area of economic game theory and then introduced into the optimization community. BLO is able to handle problems with a hierarchical structure, involving two levels of optimization tasks, where one task is nested inside the other. In machine learning and computer vision fields, despite the different motivations and mechanisms, a lot of complex problems, such as hyper-parameter optimization, multi-task and meta learning, neural architecture search, adversarial learning and deep reinforcement learning, actually all contain a series of closely related subproblms. In this paper, we first uniformly express these complex learning and vision problems from the perspective of BLO. Then we construct a best-response-based single-level reformulation and establish a unified algorithmic framework to understand and formulate mainstream gradient-based BLO methodologies, covering aspects ranging from fundamental automatic differentiation schemes to various accelerations, simplifications, extensions and their convergence and complexity properties. Last but not least, we discuss the potentials of our unified BLO framework for designing new algorithms and point out some promising directions for future research. © 2021, CC BY.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Transportation, Germs, Culture: A Dynamic Graph Model of COVID-19 Outbreak

SSRN

引用

SSRN 2020年

作者： Yang, Xiaofei Xu, Tun Jia, Peng Xia, Han Guo, Li Zhang, Lei Ye, Kai School of Computer Science and Technology Faculty of Electronic and Information Engineering Xi’an Jiaotong University Xi’an710049 China MOE Key Lab for Intelligent Networks & Networks Security Faculty of Electronic and Information Engineering Xi’an Jiaotong University Xi’an710049 China School of Automation Science and Engineering Faculty of Electronic and Information Engineering Xi’an Jiaotong University Xi’an710049 China Hugobiotech Co. Ltd. Beijing100000 China Genome Institute The First Affiliated Hospital of Xi’an Jiaotong University Xi’an710061 China The School of Life Science and Technology Xi’an Jiaotong University Xi’an710049 China China-Australia Joint Research Center for Infectious Diseases School of Public Health Xi’an Jiaotong University Health Science Center Shaanxi Xi’an710061 China Melbourne Sexual Health Centre Alfred Health Melbourne Australia Central Clinical School Faculty of Medicine Nursing and Health Sciences Monash University MelbourneVIC Australia Department of Epidemiology and Biostatistics College of Public Health Zhengzhou University Henan Zhengzhou450001 China

Background: Various forms of model have been applied to predict the trend of the epidemic since the outbreak of COVID-19 at the hardest-hit city of Wuhan. Methods: In this study, we designed a dynamic graph model, not for precisely predicting the number of infected cases, but for a glance of the dynamics under a public epidemic emergency situation and of different contributing factors. Findings: We demonstrated the impact of asymptomatic transmission in this outbreak and showed the effectiveness of city lockdown to halt virus spread within a city. We further illustrated that sudden emergence of a large number of cases could overwhelm the city medical system, and external medical aids are critical to not only containing the further spread of the virus but also reducing fatality. Interpretation: Our model simulation showed that highly populated modern cities are particularly vulnerable and lessons learned in China could facilitate other countries to plan the proactive and decisive actions. We shall pay close attention to the asymptomatic transmission being suggested by rapidly accumulating evidence as dramatic changes in quarantine protocol are required to contain SARS-CoV-2 from spreading globally. Funding Statement: This study was supported by the National Science Foundation of China (grand NO. 61702406 31671372, 31701739, and 8191101420), the National Key R&D Program of China (grand NO. 2018YFC0910400, 2017YFC0907500), the National Science and Technology Major Project of China (grand NO. 2018ZX10302205), the 'World-Class Universities and the Characteristic Development Guidance Funds for the Central Universities', Thousand Talents Plan Professorship for Young Scholars (3111500001), Xi'an Jiaotong University Young Talent Support Program and Xi’an Jiaotong University Basic Research and Profession Grant (xtr022019003). Declaration of Interests: None. © 2020, The Authors. All rights reserved.

关键词： SARS

来源：评论

学校读者我要写书评

暂无评论

Proceedings of the first workshop on weakly supervised learning (WeaSuL)

arXiv

引用

arXiv 2021年

作者： Hedderich, Michael A. Roth, Benjamin Kann, Katharina Plank, Barbara Ratner, Alex Klakow, Dietrich Madan, Vivek Khetan, Ashish Karnin, Zohar Wang, Cheng Kim, Sun Park, Taiwoo Choudhary, Sajal Park, Sunghyun Kim, Young-Bum Sarikaya, Ruhi Lee, Sungjin Huang, Xin Khetan, Ashish Cvitkovic, Milan Karnin, Zohar Cachay, Salva Rühling Boecking, Benedikt Dubrawski, Artur Tran, Linh Taghanaki, Saeid Asgari Khasahmadi, Amir Hosein Sanghi, Aditya Meyer, Matthias Thiele, Lothar Wenner, Michaela Walter, Fabian Hibert, Clément Saarland University Germany University of Vienna Austria University of Colorado Boulder United States IT University of Copenhagen Denmark University of Washington United States AWS AI Labs Amazon Alexa AI Amazon AI PostEra Technical University of Darmstadt Carnegie Mellon University Autodesk AI Lab Imperial College London Computer Engineering and Networks Laboratory ETH Zurich Zurich Switzerland Laboratory of Hydraulics Hydrology and Glaciology ETH Zurich WSL Birmensdorf Institut de Physique du Globe de Strasbourg University of Strasbourg

来源：评论

学校读者我要写书评

暂无评论

Deep reinforcement learning for fresh data collection in UAV-assisted IoT networks

arXiv

引用

arXiv 2020年

作者： Yi, Mengjie Wang, Xijun Liu, Juan Zhang, Yan Bai, Bo State Key Lab of Integrated Service Networks Information Science Institute Xidian University Xi’an Shaanxi710071 China Science and Technology on Communication Network Laboratory Shijiazhuang Hebei050081 China School of Electronics and Communication Engineering Sun Yat-sen University Guangzhou510006 China Key Laboratory of Wireless Sensor Network & Communication Shanghai Institute of Microsystem and Information Technology Chinese Academy of Sciences 865 Changning Road Shanghai200050 China School of Electrical Engineering and Computer Science Ningbo University Zhejiang315211 China 2012 Labs Huawei Technologies Co. Ltd. Hong Kong Hong Kong

Due to the flexibility and low operational cost, dispatching unmanned aerial vehicles (UAVs) to collect information from distributed sensors is expected to be a promising solution in Internet of Things (IoT), especially for time-critical applications. How to maintain the information freshness is a challenging issue. In this paper, we investigate the fresh data collection problem in UAV-assisted IoT networks. Particularly, the UAV flies towards the sensors to collect status update packets within a given duration while maintaining a non-negative residual energy. We formulate a Markov Decision Process (MDP) to find the optimal flight trajectory of the UAV and transmission scheduling of the sensors that minimizes the weighted sum of the age of information (AoI). A UAV-assisted data collection algorithm based on deep reinforcement learning (DRL) is further proposed to overcome the curse of dimensionality. Extensive simulation results demonstrate that the proposed DRL-based algorithm can significantly reduce the weighted sum of the AoI compared to other baseline algorithms. Copyright © 2020, The Authors. All rights reserved.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

IDTxl: The information dynamics toolkit xl: A python package for the efficient analysis of multivariate information dynamics in networks

arXiv

引用

arXiv 2018年

作者： Wollstadt, Patricia Lizier, Joseph T. Vicente, Raul Finn, Conor Martinez-Zarzuela, Mario Mediano, Pedro Novelli, Leonardo Wibral, Michael MEG Unit Brain Imaging Center Goethe-University Frankfurt Fankfurt am Main Germany Centre for Complex Systems Faculty of Engineering and IT University of Sydney SydneyNSW Australia Computational Neuroscience Lab Institute of Computer Science Tartu Estonia Data61 CSIRO EppingNSW Australia Communications and Signal Theory and Telematics Engineering University of Valladolid Valladolid Spain Computational Neurodynamics Group Department of Computing Imperial College London London United Kingdom Max Planck Institute for Dynamics and Self-Organization Göttingen Germany Campus Institute for Dynamics of Biological Networks Georg-August Universität Göttingen Germany

We present IDTxl (the Information Dynamics Toolkit xl), a new open source Python toolbox for effective network inference from multivariate time series using information theory, available from GitHub (https://***/pwollstadt/IDTxl). Information theory (Cover & Thomas, 2006;MacKay, 2003;Shannon, 1948) is the mathematical theory of information and its transmission over communication channels. Information theory provides quantitative measures of the information content of a single random variable (entropy) and of the information shared between two variables (mutual information). The defined measures build on probability theory and solely depend on the probability distributions of the variables involved. As a consequence, the dependence between two variables can be quantified as the information shared between them, without the need to explicitly model a specific type of dependence. Hence, mutual information is a model-free measure of dependence, which makes it a popular choice for the analysis of systems other than communication channels. Transfer entropy (TE) (Schreiber, 2000) is an extension of mutual information that measures the directed information transfer between time series of a source and a target variable. TE has become popular in many scientific disciplines to infer dependencies and whole networks from data. Notable application domains include neuroscience (Wibral, Vicente, & Lindner, 2014) and dynamical systems analysis (Lizier, Prokopenko, & Zomaya, 2014) (see Bossomaier, Barnett, Harré, & Lizier (2016) for an introduction to TE and a comprehensive discussion of its application). In the majority of the applications, TE is used in a bivariate fashion, where information transfer is quantified between all sourcetarget pairs. In a multivariate setting, however, such a bivariate analysis may infer spurious or redundant interactions, where multiple sources provide the same information about the target. Conversely, bivariate analysis may also miss synergistic

关键词： High level languages

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：