检索结果-内蒙古大学图书馆

Deep Hypercomplex Networks for Spatiotemporal Data Processing: Parameter efficiency and superior performance

IEEE SIGNAL PROCESSING MAGAZINE 2024年第3期41卷 101-112页

作者： Bojesomo, Alabi Liatsis, Panos Marzouqi, Hasan Al Khalifa Univ Sci & Technol Dept Elect Engn & Comp Sci Abu Dhabi U Arab Emirates Khalifa Univ Sci & Technol Dept Comp Sci Abu Dhabi U Arab Emirates City Univ London Elect & Elect Engn Dept London England

Hypercomplex numbers, such as quaternions and octonions, have recently gained attention because of their advantageous properties over real numbers, e.g., in the development of parameter-efficient neural networks. For instance, the 16-component sedenion has the capacity to reduce the number of network parameters by a factor of 16. Moreover, hypercomplex neural networks offer advantages in the processing of spatiotemporal data as they are able to represent variable temporal data divisions through the hypercomplex components. Similarly, they support multimodal learning, with each component representing an individual modality. In this article, the key components of deep learning in the hypercomplex domain are introduced, encompassing concatenation, activation functions, convolution, and batch normalization. The use of the backpropagation algorithm for training hypercomplex networks is discussed in the context of hypercomplex algebra. These concepts are brought together in the design of a ResNet backbone using hypercomplex convolution, which is integrated within a U-Net configuration and applied in weather and traffic forecasting problems. The results demonstrate the superior performance of hypercomplex networks compared to their real-valued counterparts, given a fixed parameter budget, highlighting their potential in spatiotemporal data processing.

关键词： Training data Convolutional neural networks Algebra Quaternions Image processing Data processing Spatiotemporal phenomena Hypercomplex Neural networks backpropagation algorithms Batch normalization

来源：评论

学校读者我要写书评

暂无评论

A ROBUST BACK-PROPAGATION LEARNING ALGORITHM FOR FUNCTION APPROXIMATION

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS 1994年第3期5卷 467-479页

作者： CHEN, DS JAIN, RC UNIV MICHIGAN DEPT ELECT ENGN & COMP SCIARTIFICIAL INTELLIGENCE LABANN ARBORMI 48109

The back propagation (BP) algorithm allows multilayer feedforward neural networks to learn input-output mappings from training samples. Due to the nonlinear modeling power of such networks, the learned mapping may interpolate all the training points. When erroneous training data are employed, the learned mapping can oscillate badly between data points. In this paper we derive a robust BP learning algorithm that is resistant to the noise effects and is capable of rejecting gross errors during the approximation process. The spirit of this algorithm comes from the pioneering work in robust statistics by Huber an d Hampel. Our work is different from that of M-estimators in two aspects: 1) the shape of the objective function changes with the iteration time;2) the parametric form of the functional approximator is no longer linear: it is a nonlinear cascade of affine transformations. In contrast to the conventional BP algorithm, three advantages of the robust BP algorithm are: 1) it approximates an underlying mapping rather than interpolating training samples;2) it is robust against gross errors;3) its rate of convergence is improved since the influence of incorrect samples is gracefully suppressed.

关键词： backpropagation algorithms Multi-layer neural network Noise robustness Neural networks Feedforward neural networks Training data Approximation algorithms Noise shaping Parametric statistics Shape

来源：评论

学校读者我要写书评

暂无评论

Fuzzy classifications using fuzzy inference networks

引用

IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS 1998年第3期28卷 334-347页

作者： Cai, LYL Kwan, HK Varian Associates Inc Ginston Res Ctr Palo Alto CA 94304 USA Univ Windsor Dept Elect Engn Windsor ON N9B 3P4 Canada

In this paper, fuzzy inference models for pattern classifications have been developed and fuzzy inference networks based on these models are proposed. Most of the existing fuzzy rule-based systems have difficulties in deriving inference rules and membership functions directly from training data, Rules and membership functions are obtained from experts. Some approaches use backpropagation (BP) type learning algorithms to learn the parameters of membership functions from training data. However, BP algorithms take a long time to converge and they require an advanced setting of the number of inference rules. The work to determine the number of inference rules demands lots of experiences from the designer. In this paper, self-organizing learning algorithms are proposed for the fuzzy inference networks. In the proposed learning algorithms, the number of inference rules and the membership functions in the inference rules will be automatically determined during the training procedure. The learning speed is fast. The proposed fuzzy inference network (FIN) classifiers possess both the structure and the learning ability of neural networks, and the fuzzy classification ability of fuzzy algorithms. Simulation results on fuzzy classification of two-dimensional data are presented and compared with those of the fuzzy ARTMAP, The proposed fuzzy inference networks perform better than the fuzzy ARTMAP and need less training samples.

关键词： Fuzzy neural networks Partitioning algorithms Inference algorithms Neural networks Pattern recognition Classification algorithms Training data Pattern classification Fuzzy systems backpropagation algorithms

来源：评论

学校读者我要写书评

暂无评论

Supervised Learning in Neural Networks: Feedback-Network-Free Implementation and Biological Plausibility

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2022年第12期33卷 7888-7898页

作者： Lin, Feng Wayne State Univ Dept Elect & Comp Engn Detroit MI 48202 USA

The well-known backpropagation learning algorithm is probably the most popular learning algorithm in artificial neural networks. It has been widely used in various applications of deep learning. The backpropagation algorithm requires a separate feedback network to back propagate errors. This feedback network must have the same topology and connection strengths (weights) as the feed-forward network. In this article, we propose a new learning algorithm that is mathematically equivalent to the backpropagation algorithm but does not require a feedback network. The elimination of the feedback network makes the implementation of the new algorithm much simpler. The elimination of the feedback network also significantly increases biological plausibility for biological neural networks to learn using the new algorithm by means of some retrograde regulatory mechanisms that may exist in neurons. This new algorithm also eliminates the need for two-phase adaptation (feed-forward phase and feedback phase). Hence, neurons can adapt asynchronously and concurrently in a way analogous to that of biological neurons.

关键词： Neurons Biological neural networks Biology backpropagation algorithms backpropagation Network topology Topology backpropagation biological plausibility deep learning neural networks

来源：评论

学校读者我要写书评

暂无评论

A MULTIPLE TIME-SCALE MODULARIZED ADAPTIVE BEAMFORMER

引用

IEEE TRANSACTIONS ON SIGNAL PROCESSING 1994年第5期42卷 1251-1256页

作者： NIEZGODA, GH KROLIK, JL DUKE UNIV DEPT ELECT ENGNDURHAMNC 27708

This paper analyzes the cascaded equivalent form of the minimum variance distortionless response (MVDR) beamformer. A unique feature of the modularized MVDR (M2VDR) beamformer is the ability to distribute adaptive degrees of freedom (DOF) among beamformer stages. This property of the M2VDR beamformer allows DOF to be adapted over different time scales. As a consequence of the multiple stage decomposition adaptive beamformer weights appear in a form which suggest an apparent increase in the number of DOF. To properly analyze this case an alternative interpretation of adaptive DOF is introduced to establish the equivalence between the MVDR beamformer and its cascaded realization. In the second part of this paper the distribution for the output power estimate of a P stage M2VDR beamformer is derived. Analysis reveals the necessity of placing constraints on the adaptation time assigned to each beamformer stage. Simulation experiments are presented that demonstrate the applicability of the M2VDR beamformer in resolving closely spaced sources under impulsive interference conditions.

关键词： Signal processing algorithms backpropagation algorithms Least squares approximation Classification algorithms Neural networks Adaptive signal processing Algorithm design and analysis System identification Speech processing Acoustic distortion

来源：评论

学校读者我要写书评

暂无评论

Particle Size Distribution Reconstruction Using a Finite Number of Its Moments through Artificial Neural Networks: A Practical Application

引用

CRYSTAL GROWTH & DESIGN 2015年第1期15卷 239-246页

作者： Cogoni, G. Frawley, P. J. Univ Limerick Dept Mech Aeronaut & Biomed Engn MSSI SSPC Castletroy Limerick Ireland

An artificial neural network (ANN) approach to reconstruct the particle size distribution (PSD) is proposed in this paper. This novel technique has been applied for acetaminophen crystallization in ethanol. Several experimental PSDs taken in different operating conditions, such as temperature and agitation degree, at different stages of the process have been considered, in order to ensure a wide range of different distributions for the system taken into account. The first stage of the ANN modeling is represented by the structure definition and the network training through a backpropagation algorithm in which the experimental PSDs and their associated vector of moments have been used. The second stage is represented by the feedforward application and validations of the proposed model estimating the PSDs using a finite set of experimental moments compared with their associated PSDs and then, using a set of time-dependent moments, obtaining the transitory PSD. The proposed approach represents a more suitable way to reconstruct the PSD in full, for the first time without assuming any reference distribution or knowing in advance the shape of the experimental PSD, leading to a generalized characterization of the PSDs with possible implementations in other multiphase unit operations.

关键词： backpropagation algorithms Particle size distribution Operating conditions Moment of force structural definition

来源：评论

学校读者我要写书评

暂无评论

CLASSIFICATION OF MULTISENSOR REMOTE-SENSING IMAGES BY STRUCTURED NEURAL NETWORKS

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 1995年第3期33卷 562-578页

作者： SERPICO, SB ROLI, F UNIV GENOA DEPT BIOPHYS & ELECTR ENGN I-16145 GENOA ITALY

This paper proposes the application of structured neural networks to classification of multisensor remote-sensing images, The purpose of our approach is to allow the interpretation of the ''network behavior,'' as it can be utilized by photointerpreters for the validation of the neural classifier. In addition, our approach gives a criterion for defining the network architecture, so avoiding the classical trial-and-error process, First of all, the architecture of structured multilayer feedforward networks is tailored to a multisensor classification problem. Then, such networks are trained to solve the problem by the error backpropagation algorithm, Finally, they are transformed into equivalent networks to obtain a simplified representation. The resulting equivalent networks may be interpreted as a hierarchical arrangement of ''committees'' that accomplish the classification task by checking on a set of explicit constraints on input data, Experimental results on a multisensor (optical and SAR) data set are described in terms of both classification accuracy and network interpretation. Comparisons with fully connected neural networks and with the kappa-nearest neighbor classifier are also made.

关键词： Remote sensing Neural networks Optical sensors Context modeling Laser radar Nonhomogeneous media backpropagation algorithms Optical computing Optical fiber networks Image sensors

来源：评论

学校读者我要写书评

暂无评论

Robust vibration control of composite beams using piezoelectric devices and neural networks

引用

SMART MATERIALS & STRUCTURES 1997年第2期6卷 178-189页

作者： Smyser, CP Chandrashekhara, K UNIV MISSOURI DEPT MECH & AEROSP ENGN & ENGN MECHROLLAMO 65409

Robust vibration control of smart composite beams using neural networks was studied. Linear quadratic Gaussian with loop transfer recovery (LQG/LTR) methodology was used to design a robust controller on the basis of the state space model of the system. The state space model of the system was obtained using the finite-element method and mode superposition. The finite-element model was based on a higher-order shear deformation theory which included the lateral strains. The mode superposition method was used to transform the coupled finite-element equations of motion in the physical coordinates into a set of reduced uncoupled equations in the modal coordinates. The performance of the LQG/LTR controller was verified for various arbitrary initial conditions. A system of neural networks was then trained to emulate the robust controller. The neural network system was trained using the backpropagation algorithm. After suitable training, the NN (neural network) controller was shown to effectively control the vibrations of the composite beam. A robustness study including the effects of tip mass, structural parameter variation, and loss of a sensor input was performed. The NN controller is shown to provide robustness and control capabilities equivalent to that of the LQG/LTR controller.

关键词： DESIGN Robust controllers State space models Equations of motion Controllers Neural network composite beams backpropagation algorithms Piezoelectric devices Robustness Mode superposition method vibration control

来源：评论

学校读者我要写书评

暂无评论

EXECUTION OF A REMOTE-SENSING APPLICATION ON A CUSTOM NEUROCOMPUTER

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS 1995年第6期6卷 1505-1515页

作者： WATKINS, SS CHAU, PM TAWEL, R LAMBRIGTSEN, B CALTECH JET PROP LABPASADENACA 91109

A radial basis function neural network was successfully applied to an area which is relatively new for neural networks: a remote sensing application that provides estimates of water vapor content, an important parameter for climate modeling. The neural network provided results which are up to 32% better than had been previously obtained using conventional statistical methods on the same data These results have implications for improved short-term weather forecasting and for long-term global climate modeling. The neural network approach is compared with the past and present operating algorithms at the National Oceanic and Atmospheric Administration, The radial basis function network's performance is compared with sigmoidal backpropagation network. Low-power electronic implementations of the neural methodology were explored to demonstrate the feasibility of placing the network on a remote sensing platform, This would permit processing the ram sensor data into information on the platform, eliminating the need to store the raw data, and helping to contain the expected explosion of climate data.

关键词： Remote sensing Neural networks Radial basis function networks Statistical analysis Weather forecasting Atmospheric modeling Predictive models backpropagation algorithms Low power electronics Explosions

来源：评论

学校读者我要写书评

暂无评论

CURVATURE-DRIVEN SMOOTHING - A LEARNING ALGORITHM FOR FEEDFORWARD NETWORKS

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS 1993年第5期4卷 882-884页

作者： BISHOP, CM Neural Computing Research Group Department of Computer Science Aston University Birmingham UK

The performance of feedforward neural networks in real applications can often be improved significantly if use is made of a priori information. For interpolation problems this prior knowledge frequently includes smoothness requirements on the network mapping, and can be imposed by the addition to the error function of suitable regularization terms. The new error function, however, now depends on the derivatives of the network mapping, and so the standard backpropagation algorithm cannot be applied. In this letter, we derive a computationally efficient learning algorithm, for a feedforward network of arbitrary topology, which can be used to minimize such error functions. Networks having a single hidden layer, for which the learning algorithm simplifies, are treated as a special case.

关键词： Smoothing methods backpropagation algorithms Training data Neural networks Interpolation Network topology Multilayer perceptrons Mean square error methods Transfer functions Feedforward neural networks

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：