检索结果-内蒙古大学图书馆

Dynamic Clustering Convolutional Neural Network 39

学校读者我要写书评

暂无评论

Dynamic Clustering Convolutional Neural Network

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Li, Tanzhe Zhang, Baochang Lyu, Jiayi Zheng, Xiawu Guo, Guodong Jin, Taisong Key Laboratory of Multimedia Trusted Perception and Effcient Computing Ministry of Education of China Xiamen University China School of Informatics Xiamen University China Hangzhou Research Institute School of Artificial Intelligence Beihang University China Nanchang Institute of Technology China School of Engineering Science University of Chinese Academy of Sciences China Ningbo Institute of Digital Twin Eastern Institute of Technology Ningbo China Key Laboratory of Oracle Bone Inscriptions Information Processing Ministry of Education of China Anyang Normal University China

ISBN: (纸本)157735897X

Convolutional neural networks (CNNs) have been playing a dominant role in computer vision. However, the existing approaches of using local window modeling in popular CNNs lack flexibility and hinder their ability to capture long-range dependencies of objects in an image. To overcome these limitations, we propose a novel CNN architecture, termed Dynamic Clustering Convolutional Neural Network (DCCNeXt). The proposed DCCNeXt takes a unique approach by employing global clustering to group image patches with similar semantics into clusters that are then convolved using the shared convolution kernels. To address the high computational complexity of global clustering, the feature vectors from each patch's subspace are extracted for efficient clustering, which makes the proposed model widely compatible with the downstream vision tasks. The extensive experiments of image classification, object detection, instance segmentation, and semantic segmentation on the benchmark datasets demonstrate that the proposed DCCNeXt outperforms the mainstream Convolutional Neural Networks (CNNs), Vision Transformers (ViTs), Vision Multi-layer Per-ceptrons (MLPs), Vision Graph Neural Networks (GNNs), and Vision Mambas (ViMs). We anticipate that this study will provide a new perspective and a promising avenue for the design of convolutional neural networks. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Convolutional neural networks

SpreadFGL: Edge-Client Collaborative Federated Graph Learning with Adaptive Neighbor Generation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhong, Luying Pi, Yueyang Chen, Zheyi Yu, Zhengxin Miao, Wang Chen, Xing Min, Geyong College of Computer and Data Science Fuzhou University China Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou University China Engineering Research Center of Big Data Intelligence Ministry of Education China School of Computing and Communications University of Lancaster United Kingdom School of Engineering Computing and Mathematics University of Plymouth United Kingdom Department of Computer Science University of Exeter United Kingdom

Federated Graph Learning (FGL) has garnered widespread attention by enabling collaborative training on multiple clients for semi-supervised classification tasks. However, most existing FGL studies do not well consider the missing inter-client topology information in real-world scenarios, causing insufficient feature aggregation of multi-hop neighbor clients during model training. Moreover, the classic FGL commonly adopts the FedAvg but neglects the high training costs when the number of clients expands, resulting in the overload of a single edge server. To address these important challenges, we propose a novel FGL framework, named SpreadFGL, to promote the information flow in edge-client collaboration and extract more generalized potential relationships between clients. In SpreadFGL, an adaptive graph imputation generator incorporated with a versatile assessor is first designed to exploit the potential links between subgraphs, without sharing raw data. Next, a new negative sampling mechanism is developed to make SpreadFGL concentrate on more refined information in downstream tasks. To facilitate load balancing at the edge layer, SpreadFGL follows a distributed training manner that enables fast model convergence. Using real-world testbed and benchmark graph datasets, extensive experiments demonstrate the effectiveness of the proposed SpreadFGL. The results show that SpreadFGL achieves higher accuracy and faster convergence against state-of-the-art algorithms. Copyright © 2024, The Authors. All rights reserved.

关键词： Graph algorithms

Collaborative learning of graph generation, clustering and classification for brain networks diagnosis

学校读者我要写书评

暂无评论

COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2022年 219卷 106772-106772页

作者： Yang, Wenju Wen, Guangqi Cao, Peng Yang, Jinzhu Zaiane, Osmar R. College of Computer Science and Engineering Northeastern University Shenyang China Key Laboratory of Intelligent Computing in Medical Image Ministry of Education Northeastern University Shenyang China Alberta Machine Intelligence Institute University of Alberta Edmonton Canada

Purpose: Accurate diagnosis of autism spectrum disorder (ASD) plays a key role in improving the condition and quality of life for patients. In this study, we mainly focus on ASD diagnosis with functional brain networks (FBNs). The major challenge for brain networks modeling is the high dimensional connectivity in brain networks and limited number of subjects, which hinders the classification capability of graph convolutional networks (GCNs). Method: To alleviate the influence of the limited data and high dimensional connectivity, we introduce a unified three-stage graph learning framework for brain network classification, involving multi-graph clustering, graph generation and graph classification. The framework combining Graph Generation, Clustering and Classification Networks (GraphCGC-Net) enhances the critical connections by multi-graph clustering (MGC) with a supervision scheme, and generates realistic brain networks by simultaneously preserving the global consistent distribution and local topology properties. Results: To demonstrate the effectiveness of our approach, we evaluate the performance of the proposed method on the Autism Brain Imaging Data Exchange (ABIDE) dataset and conduct extensive experiments on the ASD classification problem. Our proposed method achieves an average accuracy of 70.45% and an AUC of 72.76% on ABIDE. Compared with the traditional GCN model, the proposed GraphCGC-Net obtains 9.3%, and 10.64% improvement in terms of accuracy and AUC metrics, respectively. Conclusion: The comprehensive experiments demonstrate that our GraphCGC-Net is effective for graph classification in brain disorders diagnosis. Moreover, we find that MGC can generate biologically meaningful subnetworks, which is highly consistent with the previous neuroimaging-derived biomarker evidence of ASD. More importantly, the promising results suggest that applying generative adversarial networks (GANs) in brain networks to improve the classification performance is worth fur

关键词： Autism spectrum disorder Functional brain networks Multi-graph clustering Generative adversarial networks Graph convolutional networks

Identifying the turbulent flow developing inside and around the bottom trawl by Electromagnetic Current Velocity Meter approach in the flume tank

学校读者我要写书评

暂无评论

Journal of Hydrodynamics 2021年第3期33卷 636-656页

作者： Nyatchouba Nsangue Bruno Thierry Hao Tang Liu-xiong Xu Fuxiang Hu Xinxing You Micah Adekunle David Njomoue Pandong Achille College of Marine Sciences Shanghai Ocean UniversityShanghai201306China National Engineering Research Center for Oceanic Fisheries Shanghai201306China Key Laboratory of Oceanic Fisheries Exploration Ministry of Agriculture and Rural AffairsShanghai201306China The Key Laboratory of Sustainable Exploitation of Oceanic Fisheries Resources Shanghai Ocean UniversityMinistry of EducationShanghai201306China Scientific Observing and Experimental Station of Oceanic Fishery Resources Ministry of Agriculture and Rural AffairsShanghai201306China Department of Marine Biosciences Tokyo University of Marine Science and TechnologyMinatoTokyoJapan College of Fisheries and Life Sciences Shanghai Ocean UniversityShanghai201306China Faculty of Industrial Engineering University of DoualaDoualaCameroon

The Knowledge of turbulent flow developing inside and around the bottom trawl net is of great importance not only for improving the hydrodynamic performance of the gear but also for the selectivity via the fish response,such as the herding response or escape *** 3-D Electromagnetic Current Velocity Meter(ECVM)measurements were performed to investigate the effect of turbulent flow on the bottom trawl net performance and to analyze the turbulence intensity and velocity ratio inside and around different parts of the trawl *** orthogonal decomposition(POD)method was applied in order to extract the phase averaged mean velocity field of turbulent flow from each available ECVM instantaneous *** results demonstrated the existence of turbulence flow,consisting of turbulent boundary layer flow and the turbulence due to the trawl wake developing all inside and around the bottom trawl *** input streamwise velocity results in faster trawl movement and a significant turbulent *** maximum turbulence intensity inside and around trawl wing,square part,first belly,second belly,third belly,cod-end is 0.95%,1.34%,3.40%,4.10%,4.25%and 3.80%,*** was found that the mean velocity field in a turbulent flow inside and around trawl net cod-end recovered on the average was~77.58%of the input streamwise *** is~12.92%,~13.07%,~11.40%,~13.00%and~0.45%less than that inside and around trawl wing,square part,first belly,second belly,and third belly of the bottom trawl net,*** turbulent flow behavior depends strongly on the structure oscillation,input streamwise velocity and,porosity of the net *** is necessary to take into account the velocity reduction inside and around a different part of the trawl net to improve the entire drag force determination,cod-end design,and further selectivity control of the fishing gear.

关键词： Hydrodynamic bottom trawl net turbulent flow Electromagnetic Current Velocity Meter(ECVM) turbulence intensity proper orthogonal decomposition(POD)

PET: Multi-agent Independent PPO-based Automatic ECN Tuning for High-Speed Data Center Networks

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Cheng, Kai Wang, Ting Du, Xiao Du, Shuyi Cai, Haibin The Engineering Research Center of Software/Hardware Co-design Technology and Application Ministry of Education The Shanghai Key Laboratory of Trustworthy Computing Software Engineering Institute East China Normal University Shanghai200050 China The Department of Houston International Institute Dalian Maritime University Liaoning Dalian116026 China

Explicit Congestion Notification (ECN)-based congestion control schemes have been widely adopted in high-speed data center networks (DCNs), where the ECN marking threshold plays a determinant role in guaranteeing a packet lossless DCN. However, existing approaches either employ static settings with immutable thresholds that cannot be dynamically self-adjusted to adapt to network dynamics, or fail to take into account many-to-one traffic patterns and different requirements of different types of traffic, resulting in relatively poor performance. To address these problems, this paper proposes a novel learning-based automatic ECN tuning scheme, named PET, based on the multi-agent Independent Proximal Policy Optimization (IPPO) algorithm. PET dynamically adjusts ECN thresholds by fully considering pivotal congestion-contributing factors, including queue length, output data rate, output rate of ECN-marked packets, current ECN threshold, the extent of incast, and the ratio of mice and elephant flows. PET adopts the Decentralized Training and Decentralized Execution (DTDE) paradigm and combines offline and online training to accommodate network dynamics. PET is also fair and readily deployable with commodity hardware. Comprehensive experimental results demonstrate that, compared with state-of-the-art static schemes and the learning-based automatic scheme, our PET achieves better performance in terms of flow completion time, convergence rate, queue length variance, and system robustness. © 2024, CC BY.

关键词： Mammals

Evaluating Open-Domain Dialogues in Latent Space with Next Sentence Prediction and Mutual Information

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhao, Kun Yang, Bohao Lin, Chenghua Rong, Wenge Villavicencio, Aline Cui, Xiaohui Key Laboratory of Aerospace Information Security and Trusted Computing Ministry of Education School of Cyber Science and Engineering Wuhan University China Department of Computer Science The University of Sheffield United Kingdom State Key Laboratory of Software Development Environment Beihang University China School of Computer Science and Engineering Beihang University China

The long-standing one-to-many issue of the open-domain dialogues poses significant challenges for automatic evaluation methods, i.e., there may be multiple suitable responses which differ in semantics for a given conversational context. To tackle this challenge, we propose a novel learning-based automatic evaluation metric (CMN), which can robustly evaluate open-domain dialogues by augmenting Conditional Variational Autoencoders (CVAEs) with a Next Sentence Prediction (NSP) objective and employing Mutual Information (MI) to model the semantic similarity of text in the latent space. Experimental results on two open-domain dialogue datasets demonstrate the superiority of our method compared with a wide range of baselines, especially in handling responses which are distant to the golden reference responses in semantics. Copyright © 2023, The Authors. All rights reserved.

关键词： Semantics

Zero-Query Adversarial Attack on Black-box Automatic Speech Recognition Systems

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Fang, Zheng Wang, Tao Zhao, Lingchen Zhang, Shenyi Li, Bowen Ge, Yunjie Li, Qi Shen, Chao Wang, Qian Tsinghua University Beijing China Xi'an Jiaotong University Xi’an China Key Laboratory of Aerospace Information Security and Trusted Computing Ministry of Education School of Cyber Science and Engineering Wuhan University Wuhan China Institude for Network Sciences and Cyberspace & BNRist China School of Cyber Science and Engineering China

In recent years, extensive research has been conducted on the vulnerability of ASR systems, revealing that black-box adversarial example attacks pose significant threats to real-world ASR systems. However, most existing black-box attacks rely on queries to the target ASRs, which is impractical when queries are not permitted. In this paper, we propose ZQ-Attack, a transfer-based adversarial attack on ASR systems in the zero-query black-box setting. Through a comprehensive review and categorization of modern ASR technologies, we first meticulously select surrogate ASRs of diverse types to generate adversarial examples. Following this, ZQ-Attack initializes the adversarial perturbation with a scaled target command audio, rendering it relatively imperceptible while maintaining effectiveness. Subsequently, to achieve high transferability of adversarial perturbations, we propose a sequential ensemble optimization algorithm, which iteratively optimizes the adversarial perturbation on each surrogate model, leveraging collaborative information from other models. We conduct extensive experiments to evaluate ZQ-Attack. In the over-the-line setting, ZQ-Attack achieves a 100% success rate of attack (SRoA) with an average signal-to-noise ratio (SNR) of 21.91dB on 4 online speech recognition services, and attains an average SRoA of 100% and SNR of 19.67dB on 16 open-source ASRs. For commercial intelligent voice control devices, ZQ-Attack also achieves a 100% SRoA with an average SNR of 15.77dB in the over-the-air setting. Copyright © 2024, The Authors. All rights reserved.

关键词： Signal to noise ratio

Micromechanical Anisotropic Damage Approach for Understanding Tension-Induced Brittle and Ductile Behavior of Geomaterials

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Yu, Qiaojuan Zhu, Qizhi Lu, Yunxing Du, Shigui Luo, Zhanyou Yong, Rui Institute of Rock Mechanics Ningbo University Zhejiang Ningbo315211 China College of Computing Engineering Georgia Institute of Technology Atlanta30332 United States Key Laboratory of Ministry of Education for Geomechanics and Embankment Engineering Hohai University Nanjing210098 China

In this study, two anisotropic damage models, the micromechanics-based elastic anisotropic damage model (ME-ADM) and the localized elastic anisotropic damage model (LE-ADM), are proposed to investigate the ductile and brittle behaviors of anisotropic rocks under tension, based on the Mori-Tanaka homogenization method. The ME-ADM model is formulated to model ductile behavior by focusing on the diffuse damage phase and characterizing the anisotropic material as a matrix containing oriented microcracks, with the continuous damage criterion used to define the evolution of diffuse anisotropic damage due to crack growth in preferred directions. The LE-ADM model is formulated to model brittle behavior when damage transitions from the diffused phase to the localization phase, and the onset of localized cracks is determined using the Mohr’s maximization postulate method. The discontinuity damage criterion is used to describe the localized damage evolution of the crack in one preferred direction. Novel numerical algorithms are proposed to overcome the challenges associated with numerically implementing the models using a strain-controlled procedure. The numerical results of both models are compared to the derived analytical solution and verified with experimental results. The failure mode under pure tension and tensile-shear loading is analyzed and generalized based on traction and stress approaches. © 2023, The Authors. All rights reserved.

关键词： Microcracks

Joint Route Optimization and Multi-Dimensional Resource Management for Multiple Airborne Radars System in Target Tracking

学校读者我要写书评

暂无评论

Joint Route Optimization and Multi-Dimensional Resource Mana...

2021 CIE International Conference on Radar, Radar 2021

作者： Shi, Chenguang Dai, Xiangrong Salous, Sana Zhou, JianJiang Nanjing Univ. Aeronaut. Astronaut. Key Laboratory of Radar Imaging and Microwave Photonics Ministry of Education Nanjing China School of Engineering and Computing Sciences Durham University Durham United Kingdom

ISBN: (纸本)9781665498142

In this article, a joint route optimization and multi-dimensional resource management (JRO-MDRM) scheme is developed for multiple airborne radars system in target tracking application. The basis of the proposed scheme is to minimize the target tracking error and probability of intercept of the underlying system in the meantime by jointly optimizing the route planning, transmit power, dwell time, waveform bandwidth, and pulse length of each airborne radar node subject to certain system constraints. Since the corresponding variables are coupled in both objective function and constraints, and the resulting optimization model is non-linear and non-convex, an efficient solution algorithm, incorporating particle swarm optimization and cyclic minimization techniques, is put forth to deal with the above optimization problem. Simulation results illustrate the effectiveness of the developed JRO-MDRM scheme. © 2021 IEEE.

关键词： low probability of intercept (LPI) Multi-dimensional resource management multiple airborne radars system route optimization target tracking