检索结果-内蒙古大学图书馆

Fitting Multiple Machine Learning Models With Performance Based Clustering

IEEE signal processing LETTERS 2025年 32卷 816-820页

作者： Lorasdagi, Mehmet E. Koc, Ahmet B. Koc, Ali T. Kozat, Suleyman S. Bilkent Univ Dept Elect & Elect Engn TR-06800 Ankara Turkiye

Traditional machine learning approaches assume that data comes from a single generating mechanism, which may not hold for most real life data. In these cases, the single mechanism assumption can result in suboptimal performance. We introduce a clustering framework that eliminates this assumption by grouping the data according to the relations between the features and the target values, and we obtain multiple separate models to learn different parts of the data. We further extend our framework to applications having streaming data where we produce outcomes using an ensemble of models. For this, the ensemble weights are updated based on the incoming data batches. We demonstrate the performance of our approach over the widely-studied real life datasets, showing significant improvements over the traditional single-model approaches.

关键词： Data models Clustering algorithms Vectors signal processing algorithms Wind turbines Training Predictive models Data mining Convergence Wind speed Clustering expectation-maximization streaming data ensemble methods

来源：评论

学校读者我要写书评

暂无评论

Augmented Statistics of Quaternion Random Variables: A lynchpin of quaternion learning machines

引用

IEEE signal processing MAGAZINE 2024年第3期41卷 72-87页

作者： Took, Clive Cheong Talebi, Sayed Pouria Alcala, Rosa Maria Fernandez Mandic, Danilo P. Royal Holloway Univ London Egham TW20 0EX England Univ Roehampton London SW15 5PH England Univ Jaen Dept Stat & Operat Res Jaen 23071 Spain Imperial Coll London SW7 2BT England

Learning machines for vector sensor data are naturally developed in the quaternion domain and are underpinned by quaternion statistics. To this end, we revisit the "augmented" representation basis for discrete quaternion random variables (RVs) q(a)[n], i.e., [q [n] q(t)[n] q(j) [n] q(k) [n]], and demonstrate its pivotal role in the treatment of the generality of quaternion RVs. This is achieved by a rigorous consideration of the augmented quaternion RV and by involving the additional second-order statistics, besides the traditional covariance E{q[n]q*[n]} [1]. To illuminate the usefulness of quaternions, we consider their most well-known application-3D orientation-and offer an account of augmented statistics for purely imaginary (pure) quaternions. The quaternion statistics presented here can be exploited in the analysis of existing and the development of novel statistical machine learning methods, hence acting as a lynchpin for quaternion learning machines.

关键词： Machine learning algorithms Three-dimensional displays Quaternions Image processing signal processing algorithms Tutorials Machine learning Hypercomplex

来源：评论

学校读者我要写书评

暂无评论

Ridge Detection for Nonstationary Multicomponent signals With Time-Varying Wave-Shape Functions and its Applications

引用

IEEE TRANSACTIONS ON signal processing 2024年 72卷 4843-4854页

作者： Su, Yan-Wei Liu, Gi-Ren Sheu, Yuan-Chung Wu, Hau-Tieng Natl Yang Ming Chiao Tung Univ Dept Appl Math Hsinchu 300093 Taiwan Natl Cheng Kung Univ Dept Math Tainan 70101 Taiwan Natl Taiwan Univ Natl Ctr Theoret Sci Taipei 10617 Taiwan NYU Courant Inst Math Sci New York NY 10012 USA

We introduce a novel ridge detection algorithm for time-frequency (TF) analysis, particularly tailored for intricate nonstationary time series encompassing multiple non-sinusoidal oscillatory components. The algorithm is rooted in the distinctive geometric patterns that emerge in the TF domain due to such non-sinusoidal oscillations. We term this method shape-adaptive mode decomposition-based multiple harmonic ridge detection (SAMD-MHRD). A swift implementation is available when supplementary information is at hand. We demonstrate the practical utility of SAMD-MHRD through its application to a real-world challenge. We employ it to devise a cutting-edge walking activity detection algorithm, leveraging accelerometer signals from an inertial measurement unit across diverse body locations of a moving subject.

关键词： Harmonic analysis signal processing algorithms Mathematical models Time-frequency analysis Accelerometers Time series analysis Spectrogram Feature extraction Oscillators Noise reduction ridge detection curve extraction wave-shape function synchrosqueezing transform

来源：评论

学校读者我要写书评

暂无评论

Urban SAR Tomography With Z-Structure Constraint of Buildings

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2024年 17卷 9945-9960页

作者： Guo, Rui Gao, Yuxin Ren, Zishuai Zhang, Zhao Xie, Jinwei Yang, Bei Northwestern Polytech Univ Sch Automat Xian 710072 Peoples R China Jiangsu Shuguang Electroopt Co Ltd Yangzhou 225009 Peoples R China ZEEKR Automobile Co Ltd Shanghai 200232 Peoples R China Nanjing Res Inst Elect Technol Nanjing 210039 Peoples R China

Synthetic aperture radar tomography (TomoSAR) technology effectively mitigates issues such as severe layover in high-resolution synthetic aperture radar (SAR) urban imagery, presenting numerous applications in reconstructing complex 3-D urban scenes. However, prevailing TomoSAR methodologies usually overlook the geometrical feature of the targets, which are particularly pronounced in buildings in urban scenes. To improve the quality of 3-D reconstruction of building targets for large-scale urban scenes, we propose geometric-guided TomoSAR processing in this article. First, the geometrical feature referred to a Z-structure of buildings in TomoSAR point cloud is studied. Then, a Z-structure information extraction approach is proposed to provide prior information for subsequent geometric constraints. Finally, the Z-structure-constraint-based tomographic algorithm is proposed, optimizing the solution space of the original algorithms. Experiments are conducted on real SAR data, and 3-D point clouds of the entire urban scenes are obtained. The proposed algorithm demonstrates commendable performance in point cloud entropy and concentration, especially in the inversion of different floor structures.

关键词： Buildings Point cloud compression Three-dimensional displays Imaging Tomography Image reconstruction signal processing algorithms Compressed sensing (CS) geometrical feature synthetic aperture radar tomography (TomoSAR) 3-D reconstruction of buildings

来源：评论

学校读者我要写书评

暂无评论

Quaternion-Based Arithmetic in Quantum Information processing: A promising approach for efficient color quantum imaging

引用

IEEE signal processing MAGAZINE 2024年第2期41卷 64-74页

作者： Grigoryan, Artyom M. Agaian, Sos S. Univ Texas San Antonio Dept Elect Engn San Antonio TX 78249 USA CUNY Coll Staten Isl Dept Comp Sci Staten Isl NY USA

Classical color image processing, image recognition, and machine learning introduce nonlinearity, causing the collapse of the quantum state into classical probability perceptrons after measurements, due to the inherent linearity of quantum computing. To address this challenge, quaternion-based arithmetic offers a promising approach. By treating the primary color components as a single unit using quaternion algebra, nonlinear relationships can be implemented, effectively manipulating higher-dimensional color data. This article aims to achieve efficient and accurate color quantum image processing (QIP) by introducing new quaternion quantum-based color imaging tools based on multiplicative arithmetic on two-qubits and quantum superpositions. The approach includes the concept of a quaternion Fourier transform (QFT) in two-qubit-based color image representation. To end, we discuss possible applications of the proposed methods in color quantum imaging.

关键词： Quantum computing Image color analysis Algebra Visual communication Quaternions signal processing algorithms Color

来源：评论

学校读者我要写书评

暂无评论

DHRL-FNMR: An Intelligent Multicast Routing Approach Based on Deep Hierarchical Reinforcement Learning in SDN

引用

IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT 2024年第5期21卷 5733-5755页

作者： Ye, Miao Zhao, Chenwei Wen, Peng Wang, Yong Wang, Xiaoli Qiu, Hongbing Guilin Univ Elect Technol Sch Informat & Commun Guilin 541004 Peoples R China Guilin Univ Elect Technol Sch Comp & Informat Secur Guilin 541004 Peoples R China Xidian Univ Sch Comp Sci & Technol Xian 710071 Shaanxi Peoples R China

The multicast routing problem in software-defined networking (SDN) is an NP-hard problem. The existing solution methods based on deep strength learning suffer from the problems of branch redundancy, an excessively large action space and slow convergence of the intelligent models. In this paper, an intelligent multicast routing algorithm based on deep hierarchical reinforcement learning is proposed to circumvent the aforementioned problems. First, the optimal multicast tree problem is decomposed into two subproblems: fork node selection and the construction of an optimal path from a fork node to a destination node. Second, a multichannel matrix is designed as the state space for the internal and external controllers of hierarchical reinforcement learning based on the global network-aware information characteristics of SDN. Then, different action spaces are designed for the upper and lower subproblems, four action selection policies are designed for constructing multicast paths, and different reward policies are designed at different levels. Finally, a series of experiments and their results show that the designed algorithm not only searches the multicast tree efficiently but also converges faster and without redundant branches, with better performance in terms of bandwidth, delay and packet loss rate than the current mainstream solution algorithms. The codes for DHRL-FNMR are open and available at https://***/GuetYe/DHRL-FNMR.

关键词： Multicast algorithms Routing Aerospace electronics Heuristic algorithms signal processing algorithms Bandwidth Reinforcement learning Deep hierarchical reinforcement learning multicast tree deep reinforcement learning software-defined networking

来源：评论

学校读者我要写书评

暂无评论

Hypernetwork Based Model-Driven Channel Neural Decoding

引用

IEEE ACCESS 2024年 12卷 73228-73237页

作者： Liang, Yuanhui Lam, Chan-Tong Wu, Qingle Ng, Benjamin K. Im, Sio-Kei Macao Polytech Univ Fac Appl Sci Macau Peoples R China Sichuan Univ Sci & Engn Sch Automat & Informat Engn Yibin 644000 Peoples R China Artificial Intelligence Key Lab Sichuan Prov Yibin 644000 Peoples R China Macao Polytech Univ Macau Peoples R China

Channel decoding algorithms based on model-driven deep learning, also known as channel neural decoding algorithms, have received a lot of attention in recent years. However, the internal parameters and number of layers of the current channel neural decoding algorithm cannot be changed after training. Once changed, retraining of the channel neural decoding network is required. Hypernetwork is a neural network that can generate internal parameters for the main neural network to reduce the training cost of the main neural network and improve the flexibility of the main neural network. In this study, a novel hypernetwork based channel neural decoder is proposed for neural belief propagation algorithms (NBP), including the neural normalized min-sum (NNMS) and neural offset min-sum (NOMS) algorithms. According to the type of information interaction between the hypernetwork and the main decoding network, hypernetwork-based channel neural decoders can be divided into two types: static and dynamic. The internal parameters of the static hypernetwork-based channel neural decoder can be updated as needed without retraining of the main network. In addition to this benefit, the number of layers of the dynamic hypernetwork-based channel neural decoder can also be adjusted. Experimental results show that, compared with the existing NNMS decoding algorithms, the proposed hypernetwork-based NNMS decoding algorithms can achieve better performance on both low-density parity-check (LDPC) and Bose-Chaudhuri-Hocquenghem (BCH) codes.

关键词： Decoding Artificial neural networks Neural networks Heuristic algorithms Training signal processing algorithms Deep learning Modeling Parity check codes Model-driven hypernetwork channel neural decoding LDPC codes BCH codes

来源：评论

学校读者我要写书评

暂无评论

Cluster Head Detection for Hierarchical UAV Swarm With Graph Self-Supervised Learning

引用

IEEE TRANSACTIONS ON signal processing 2024年 72卷 5517-5532页

作者： Mou, Zhiyu Gao, Feifei Liu, Jun Yun, Xiang Wu, Qihui Tsinghua Univ THUAI Inst Artificial Intelligence Beijing 100084 Peoples R China Tsinghua Univ State Key Lab Intelligence Technol & Syst Beijing 100084 Peoples R China Tsinghua Univ Beijing Natl Res Ctr Informat Sci & Technol BNRist Beijing 100084 Peoples R China Tsinghua Univ Sch Informat Sci & Technol Dept Automat Beijing 100084 Peoples R China Tsinghua Univ Inst Network Sci & Cyberspace Beijing 100084 Peoples R China Baicells Technol Co Ltd Beijing 100044 Peoples R China Nanjing Univ Aeronaut & Astronaut Coll Elect & Informat Engn Nanjing 210016 Peoples R China

In this paper, we study the cluster head detection problem of a hierarchical unmanned aerial vehicle (UAV) swarm network (USNET), consisting of multiple double-level clusters, from the perspective of the observers who are assumed not to know any internal information on the USNET. Specifically, we propose a multi-cluster graph attention self-supervised learning (MC-GASSL) framework for the observers. The MC-GASSL comprises two main algorithms, including the sequential metric learning (SML) algorithm for detecting the clusters in the USNET, and the graph attention self-supervised learning (GASSL) algorithm for detecting the cluster head in each cluster. The basic idea of the SML is to learn an embedding space where the UAVs can be easily clustered based on Euclidean distances, and the GASSL works by calculating the attention weights between UAVs and finding the attention center of the cluster. Extensive numerical results validate the effectiveness of the MC-GASSL. Moreover, the ablation study shows that the GASSL can detect the cluster head in single clusters with over 98% detection rate on average. The ablation study also shows that the clustering purity of the USNET with the SML exceeds that with traditional clustering algorithms by at least 10% on average.

关键词： Autonomous aerial vehicles Observers Clustering algorithms Self-supervised learning signal processing algorithms Measurement Indexes Cluster head detection graph attention network self-supervised learning hierarchical UAV swarm

来源：评论

学校读者我要写书评

暂无评论

On the Impact of Random Node Sampling on Adaptive Diffusion Networks

引用

IEEE TRANSACTIONS ON signal processing 2024年 72卷 3973-3989页

作者： Tiglea, Daniel G. Candido, Renato Silva, Magno T. M. Univ Sao Paulo Elect Syst Engn Dept Escola Politecn BR-05508010 Sao Paulo Brazil

In this paper, we analyze the effects of random sampling on adaptive diffusion networks. These networks consist in a collection of nodes that can measure and process data, and that can communicate with each other to pursue a common goal of estimating an unknown system. In particular, we consider in our theoretical analysis the diffusion least-mean-squares algorithm in a scenario in which the nodes are randomly sampled. Hence, each node may or may not adapt its local estimate at a certain iteration. Our model shows that a reduction in the sampling probability leads to a noticeable deterioration in the convergence rate, and, if the nodes cooperate, to a slight decrease in the steady-state Network Mean-Square Deviation (NMSD), assuming that the environment is stationary and that all other parameters of the algorithm are kept fixed. Furthermore, we also investigate the effects of the random node sampling on the network stability.

关键词： signal processing algorithms Adaptive systems Vectors Convergence Steady-state Simulation Transient analysis Adaptive diffusion networks distributed signal processing sampling stability asynchronous networks

来源：评论

学校读者我要写书评

暂无评论

Switching Games for Image Compression

引用

IEEE signal processing LETTERS 2025年 32卷 1016-1020页

作者： Huhtanen, Marko Univ Oulu Fac Informat Technol & Elect Engn Oulu 090570 Finland

To compress an image, a technique based on optimal scalings with diagonal matrices is described. To start the process, an initial image of high compression ratio is required. Such an image can be produced, for example, with the 2D FFT or 2D DCT of the original image. Principal component analysis is a special case of this compression technique where the initial image is extremely rough, consisting of the first 2D Fourier basis function only. This initial image is then optimally orthogonalized and expanded by iteratively applying diagonal matrices from the left and right to attain double orthogonality. The process can be viewed as a continuous version of Berlekamp's switching game.

关键词： Image coding Switches Principal component analysis Discrete cosine transforms Games Eigenvalues and eigenfunctions Approximation algorithms Solids signal processing algorithms Matrix decomposition Lossy compression double orthogonality principal component analysis 2D FFT 2D DCT

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：