检索结果-内蒙古大学图书馆

SSRN 2022年

作者： Bouchoucha, Rached Ben Braiek, Houssem Khomh, Foutse Bouzidi, Sonia Zaatour, Rania Department of Computer Engineering and Software Engineering Polytechnique MontrealQCH3T 1J4 Canada Department of Computer Engineering and Software Engineering National Institute of Applied Science and Technology Tunis1080 Tunisia

Remote sensing has proven its utility in many critical domains, such as medicine, military, and ecology. Recently, we have been witnessing a surge in the adoption of deep learning (DL) techniques by the remote sensing community. DL-based classifiers, such as convolutional neural networks (CNNs), have been reported to achieve impressive predictive performances reaching 99\% of accuracy when applied to hyperspectral images (HSIs), a high-dimensional type of remote sensing data. However, these deep learners are known to be highly sensitive to even slight perturbations of their high-dimensional raw inputs. In real-world contexts, concerns can be raised about how robust they really are against corner-case scenarios. When HSI classifiers are applied in safety-critical applications, ensuring an adequate level of robustness is crucial to prevent unexpected system behaviors. Yet, there are few studies dealing with their robustness, nor are RGB-testing methods able to cover the HSI-specific challenges. This led us to propose a systematic testing method to assess the robustness of the CNNs trained to classify HSIs. First, we elaborate domain-specific metamorphic transformations that simulate naturally-occurring distortions of remote sensing HSIs. Then, we leverage metaheuristic search algorithms to optimize the fitness of synthetically-distorted inputs to stress the weaknesses of the on-testing CNN, while remaining in compliance with domain expert requirements, in order to preserve the semantic of the generated inputs. Relying on our metamorphic testing method, we assess the robustness of established and novel CNNs for HSI classification, and demonstrate their failure, on average, in 25\% of the produced test cases. Furthermore, we fine-tuned the tested CNNs on training data augmented with these failure-revealing metamorphic transformations. Results show that the fined-tuning successfully fixed at least 90\% of the CNN weaknesses, with less than 1\% of degradation in the origi

关键词： Remote sensing

来源：评论

学校读者我要写书评

暂无评论

Trustworthy Model Elements as Basis for Seamless Bim Collaboration

SSRN

引用

SSRN 2024年

作者： Paskaleva, Galina Mazak-Huemer, Alexandra Huemer, Christian TU Wien Institute of Information Systems Engineering Favoritenstr. 9-11 ViennaA-1040 Austria Johannes Kepler University Linz Department of Business Informatics - Software Engineering Altenberger Strasse 69 Science Park 3 LinzA-4040 Austria

Current applications of BIM in the Architecture engineering and Construction (AEC) industry require multiple stakeholders to collaborate on the same digital model over multiple project phases, sharing the responsibility for the same model element. This requires trust, both between the stakeholders and between each stakeholder and the BIM technology. Lack of trust often manifests as poor collaboration, model duplications and break-down of model continuity. In this work, we present a method for the introduction of trust into the digital twin of a built structure by creating trustworthy model elements. The source of trust can be any technical or legal requirement. Such trusted model elements can offer a formal guarantee that the corresponding requirement has been fulfilled. Furthermore, this guarantee can be renewed automatically after each update by any project stakeholder. Finally, we introduce a standardized workflow for the maintenance and update of already established trusted elements in a digital twin. © 2024, The Authors. All rights reserved.

关键词： Architectural design

来源：评论

学校读者我要写书评

暂无评论

LCM: Locally Constrained Compact Point Cloud Model for Masked Point Modeling

arXiv

引用

arXiv 2024年

作者： Zha, Yaohua Li, Naiqi Wang, Yanzi Dai, Tao Guo, Hang Chen, Bin Wang, Zhi Ouyang, Zhihao Xia, Shu-Tao Tsinghua Shenzhen International Graduate School Tsinghua University China Institute of Visual Intelligence Pengcheng Laboratory China College of Computer Science and Software Engineering Shenzhen University China Harbin Institute of Technology Shenzhen China Bytedance Inc China

The pre-trained point cloud model based on Masked Point Modeling (MPM) has exhibited substantial improvements across various tasks. However, these models heavily rely on the Transformer, leading to quadratic complexity and limited decoder, hindering their practice application. To address this limitation, we first conduct a comprehensive analysis of existing Transformer-based MPM, emphasizing the idea that redundancy reduction is crucial for point cloud analysis. To this end, we propose a Locally constrained Compact point cloud Model (LCM) consisting of a locally constrained compact encoder and a locally constrained Mamba-based decoder. Our encoder replaces self-attention with our local aggregation layers to achieve an elegant balance between performance and efficiency. Considering the varying information density between masked and unmasked patches in the decoder inputs of MPM, we introduce a locally constrained Mamba-based decoder. This decoder ensures linear complexity while maximizing the perception of point cloud geometry information from unmasked patches with higher information density. Extensive experimental results show that our compact model significantly surpasses existing Transformer-based models in both performance and efficiency, especially our LCM-based Point-MAE model, compared to the Transformer-based model, achieved an improvement of 1.84%, 0.67%, and 0.60% in average accuracy on the three variants of ScanObjectNN while reducing parameters by 88% and computation by 73%. Code is available at https://***/zyh16143998882/LCM. Copyright © 2024, The Authors. All rights reserved.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

MV-MOS: Multi-View Feature Fusion for 3D Moving Object Segmentation

MV-MOS: Multi-View Feature Fusion for 3D Moving Object Segme...

引用

IEEE International Conference on Robotics and Biomimetics

作者： Jintao Cheng Xingming Chen Jinxin Liang Xiaoyu Tang Xieyuanli Chen Dachuan Li Department of Computer Science and Engineering Research Institute of Trustworthy Autonomous Systems Southern University of Science and Technology Shenzhen China School of Electronic and Information Engineering South China Normal University Foshan China College of Intelligence Science and Technology National University of Defense Technology Changsha China

ISBN: (数字)9798331509644

ISBN: (纸本)9798331509651

Effectively summarizing dense 3D point cloud data and extracting motion information of moving objects (moving object segmentation, MOS) is crucial to autonomous driving and robotics applications. How to effectively utilize motion and semantic features and avoid information loss during 3D-to-2D projection is still a key challenge. In this paper, we propose a novel multi-view MOS model (MV-MOS) by fusing motion-semantic features from different 2D representations of point clouds. To effectively exploit complementary information, the motion branches of the proposed model combines motion features from both bird's eye view (BEV) and range view (RV) representations. In addition, a semantic branch is introduced to provide supplementary semantic features of moving objects. Finally, a Mamba module is utilized to fuse the semantic features with motion features and provide effective guidance for the motion branches. We validated the effectiveness of the proposed multi-branch fusion MOS framework via comprehensive experiments, and our proposed model outperforms existing state-of-the-art models on the SemanticKITTI benchmark. The implementation codes are available at https://***/Chengjt1999/MV-MOS.

关键词： Point cloud compression Solid modeling Three-dimensional displays Computational modeling Biological system modeling Semantics Neural networks Object segmentation Robots Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

On Improved Commutation for Moving-Magnet Planar Actuators

arXiv

引用

arXiv 2023年

作者： Broens, Yorick Butler, Hans Tóth, Roland The Department of Electrical Engineering Eindhoven University of Technology Eindhoven Netherlands ASML Veldhoven Netherlands The Systems and Control Laboratory Institute for Computer Science and Control Hungary

The demand for high-precision and high-throughput motion control systems has increased significantly in recent years. The use of moving-magnet planar actuators (MMPAs) is gaining popularity due to their advantageous characteristics, such as complete environmental decoupling and reduction of stage mass. Nonetheless, model-based commutation techniques for MMPAs are compromised by misalignment between the mover and coil array and mismatch between the ideal electromagnetic model and the physical system, often leading to decreased system performance. To address this issue, a novel improved commutation approach is proposed in this paper, which is applicable for general planar motor applications, by means of dynamic regulation of the position dependence of the ideal model-based commutation algorithm, which allows for attenuation of magnetic misalignment, manufacturing inaccuracies and other unmodelled phenomena. The effectiveness of the proposed approach is validated through experiments using a state-of-the-art moving-magnet planar actuator *** Codes 37M15 © 2023, CC BY.

关键词： Alignment

来源：评论

学校读者我要写书评

暂无评论

Neurocomputational Modelling of EEG Connectivity: Links Between Depression, Inflammation, and Gut-Microbiome 17th

Neurocomputational Modelling of EEG Connectivity: Links Betw...

引用

17th International Conference on Brain Informatics, BI 2024

作者： Doborjeh, Zohreh lavin, Daniel Hunter, Kirsty Heym, Nadja Heasman, Bryony Doborjeh, Maryam Kasabov, Nikola Gibson, Glen Sumich, Alexander School of Engineering Computer and Mathematical Sciences Knowledge Engineering and Discovery Research Institute Auckland University of Technology Auckland New Zealand School of Population Health Faculty of Medical and Health Sciences The University of Auckland Auckland New Zealand Department of Sport Science School of Science and Technology Nottingham Trent University Nottingham United Kingdom NTU Psychology School of Social Sciences Nottingham Trent University Nottingham United Kingdom School of Software Engineering Dalian University Dalian China Institute for Information and Communication Technologies Bulgarian Academy of Sciences Sofia Bulgaria Department of Food and Nutritional Sciences The University of Reading Whiteknights Reading United Kingdom Department of Psychology and Neuroscience Auckland University of Technology Auckland New Zealand

ISBN: (纸本)9789819632930

The immune and enteric systems have been implicated in psychopathology, including depressive disorders. However, the precise neurocognitive mechanisms remain unclear. The present study uses Brain-Inspired Spiking Neural Network (SNN) technique to model electroencephalographic (EEG) data as a function of gut-microbiota, inflammation, and depressive symptoms. Forty-two participants (nonclinical cohort) were assessed for Bacteroides-Prevotella (BAC) colonization in faecal samples, Interleukin 6 (IL-6) in peripheral blood, depressive symptoms (self-report of cognitive and non-cognitive symptoms) and brain function (resting state EEG). SNN models were applied to EEG data to generate connection weight values across scalp regions. Connection weights were visualized as a function of 1) depression;2) inflammation;3) gut microbiome. Lower depression and higher BAC were associated with higher frontal connection weights (cognitive depression = bilateral, non-cognitive and BAC = left hemisphere only). Lower cognitive depression and lower IL-6 were associated with higher posterior connection weights. The SNN models achieved higher accuracy for theta and alpha EEG sub-bands, highlighting their relevance in differentiating between groups. These findings emphasize the potential of SNNs in predicting parameters from brain functional connectivity data and developing targeted diagnostic and intervention strategies. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Two-Player Incomplete Games of Resilient Multiagent systems

引用

IFAC-PapersOnLine 2023年第2期56卷 258-263页

作者： Yurid E. Nugraha Tomohisa Hayakawa Hideaki Ishii Ahmet Cetinkaya Quanyan Zhu Department of Electrical Engineering Institut Teknologi Sepuluh Nopember Surabaya 60111 Indonesia Department of Systems and Control Engineering Tokyo Institute of Technology Tokyo 152-8552 Japan Department of Computer Science Tokyo Insitute of Technology Yokohama 226-8502 Japan Department of Functional Control Systems Shibaura Institute of Technology Tokyo 135-8548 Japan Department of Electrical and Computer Engineering New York University Brooklyn NY 11201 USA

Evolution of agents’ dynamics of multiagent systems under consensus protocol in the face of jamming attacks is discussed, where centralized parties are able to influence the control signals of the agents. In this paper we focus on a game-theoretical approach of multiagent systems where the players have incomplete information on their opponents’ strength. We consider repeated games with both simultaneous and sequential player actions where players update their beliefs of each other over time. The effect of the players’ optimal strategies according to Bayesian Nash Equilibrium and Perfect Bayesian Equilibrium on agents’ consensus is examined. It is shown that an attacker with incomplete knowledge may fail to prevent consensus despite having sufficient resources to do so.

关键词：

来源：评论

学校读者我要写书评

暂无评论

The Gate Control Method by Using Reinforcement Learning in Time-Sensitive Networking 10

The Gate Control Method by Using Reinforcement Learning in T...

引用

10th International Conference on Applied System Innovation, ICASI 2024

作者： Liu, Cheng-En Huang, Shih-Yun Chang, Yao-Chung Lin, Yu-Shan Chao, Han-Chieh Tunghai University Department of Computer Science No. 1727 Sec. 4 Taiwan Boulevard Xitun District Taichung City Taiwan National Taitung University Department of Computer Science and Information Engineering 369 Sec. 2 University Rd. Taitung Taiwan National Taitung University Department of Information Science and Management Systems 369 Sec. 2 University Rd. Taitung Taiwan Tamkang University Department of Artificial Intelligence No.151 Yingzhuan Rd. Tamsui Dist. New Taipei City Taiwan National Dong Hwa University Department of Electrical Engineering No. 1 Sec. 2 Da Hsueh Rd. Shoufeng Hualien Taiwan Institute of Computer Science and Innovation Ucsi University No.1 Jalan UCSI UCSI Heights Cheras Kuala56000 Malaysia

ISBN: (纸本)9798350394924

The time-sensitive networking (TSN) is one of the critical technologies for the future development of the industrial Internet of Things (IIoT). This is because TSN can ensure the delay and reliability of network transmission. On the other hand, TSN has also introduced the IEEE 802.1 Qbv. Since the scheduling of IEEE 802.1 Qbv is based on time control gates, the time synchronization (TAS) function can also allow TSN to communicate with fifth-generation wireless communication (5G) to support IEEE 802.1 Qbv. Because IEEE 802.1 Qbv will affect delay and jitter, it is one of the issues that urgently needs to be solved. Considering the problem of time synchronization, this paper focuses on the issue of queue gate control and proposes reinforcement learning (RL) to help us. Since reinforcement learning does not require over-examining the characteristics of past data but depends on environmental changes, adjustment will be more convenient. In addition, this paper also carefully explores the queuing status and defines in detail the problem to be solved. Finally, the experimental results show that the proposed gate control using RL is more efficient than other comparison methods. © 2024 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

DSD-MatchingNet:Deformable sparse-to-dense feature matching for learning accurate correspondences

引用

Virtual Reality & Intelligent Hardware 2022年第5期4卷 432-443页

作者： Yicheng ZHAO Han ZHANG Ping LU Ping LI Enhua WU Bin SHENG Department of Computer Science and Engineering Shanghai Jiao Tong UniversityShanghai 200240China ZTE Corporation Shenzhen 518057China State Key Laboratory of Mobile Network and Mobile Multimedia Technology Shenzhen 518057China Department of Computing The Hong Kong Polytechnic UniversityHong Kong 999077China School of Design The Hong Kong Polytechnic UniversityHong Kong 999077China State Key Laboratory of Computer Science Institute of SoftwareChinese Academy of SciencesBeijing 100190China Faculty of Science and Technology University of MacaoMacao 999078China

Background Exploring correspondences across multiview images is the basis of various computer vision ***,most existing methods have limited accuracy under challenging *** To learn more robust and accurate correspondences,we propose DSD-MatchingNet for local feature matching in this ***,we develop a deformable feature extraction module to obtain multilevel feature maps,which harvest contextual information from dynamic receptive *** dynamic receptive fields provided by the deformable convolution network ensure that our method obtains dense and robust ***,we utilize sparse-to-dense matching with symmetry of correspondence to implement accurate pixel-level matching,which enables our method to produce more accurate *** Experiments show that our proposed DSD-MatchingNet achieves a better performance on the image matching benchmark,as well as on the visual localization ***,our method achieved 91.3%mean matching accuracy on the HPatches dataset and 99.3%visual localization recalls on the Aachen Day-Night dataset.

关键词： Image matching Deformable convolution network Sparse-to-dense matching

来源：评论

学校读者我要写书评

暂无评论

Improved distributed quantum algorithm for Simon’s problem

arXiv

引用

arXiv 2025年

作者： Li, Hao Qiu, Daowen Qiu, D. Institute of Quantum Computing and Software School of Computer Science and Engineering Sun Yat-sen University Guangzhou510006 China The Guangdong Key Laboratory of Information Security Technology Sun Yat-sen University Guangzhou510006 China

Simon’s problem is one of the most important problems demonstrating the power of quantum computing. Recently, an interesting distributed quantum algorithm for Simon’s problem was proposed, where a key sorting operator requiring a large number of qubits was employed. In this paper, we design an improved distributed quantum algorithm for Simon’s problem without using sorting operators, and our algorithm has the advantage of reducing half number of qubits required for a single computing node. Moreover, our algorithm does not involve the classical search process. Copyright © 2025, The Authors. All rights reserved.

关键词： Qubits

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：