检索结果-内蒙古大学图书馆

23rd IEEE International Conference on Machine Learning and Applications, ICMLA 2024

作者： Adigun, Olaoluwa Kosko, Bart Signal and Image Processing Institute Department of Electrical and Computer Engineering Los AngelesCA90089-2564 United States

ISBN: (纸本)9798350374889

We introduce new soft diamond regularizers that both improve synaptic sparsity and maintain classification accuracy in deep neural networks. These parametrized regularizers outperform the state-of-the-art hard-diamond Laplacian regularizer of Lasso regression and classification. They use thick-tailed symmetric alpha-stable (Sa S) bell-curve synaptic weight priors that are not Gaussian and so have thicker tails. The geometry of the diamond-shaped constraint set varies from a circle to a star depending on the tail thickness and dispersion of the prior probability density function. Training directly with these priors is computationally intensive because almost all Sa S probability densities lack a closed form. A precomputed lookup table removed this computational bottleneck. We tested the new soft diamond regularizers with deep neural classifiers on the three datasets CIFAR-10, CIFAR-100, and Caltech-256. The regularizers improved the accuracy of the classifiers. The improvements included 4.57% on CIFAR-10, 4.27% on CIFAR-100, and 6.69% on Caltech-256. They also outperformed L2 regularizers on all the test cases. Soft diamond regularizers also outperformed L1 lasso or Laplace regularizers because they better increased sparsity while improving classification accuracy. Soft-diamond priors substantially improved accuracy on CIFAR-10 when combined with dropout, batch, or data-augmentation regularization. © 2024 IEEE.

关键词： Probability density function

来源：评论

学校读者我要写书评

暂无评论

Bidirectional Backpropagation Autoencoding Networks for image Compression and Denoising 22

Bidirectional Backpropagation Autoencoding Networks for Imag...

引用

22nd IEEE International Conference on Machine Learning and Applications, ICMLA 2023

作者： Adigun, Olaoluwa Kosko, Bart Signal and Image Processing Institute Department of Electrical and Computer Engineering Los AngelesCA90089-2564 United States

ISBN: (纸本)9798350345346

A bidirectional autoencoder learns or approximates an identity mapping as it trains a single network with a version of the new bidirectional backpropagation algorithm. Ordinary unidirectional autoencoders find many uses in image processing and in large language models. But they use separate networks for encoding and decoding. Bidirectional auto encoders use the same synaptic weights for encoding and decoding. The forward pass encodes while the backward pass decodes. Bidirectional auto encoders improved network performance and significantly reduced memory usage and used fewer parameters. Simulations compared unidirectional with bidirectional autoencoders for image compression and de noising. The models trained on the MNIST handwritten-digit and CIFAR-IO image datasets. The performance measures were the peak signal-to-noise ratio and the index of structural similarity. Bidirectional autoencoders outperformed unidirectional autoencoders and still reduced the number of trainable synaptic parameters by about 50%. © 2023 IEEE.

关键词： autoencoders bidirectional au-toencoders Bidirectional backpropagation bidirectional networks large language models variational autoencoders

来源：评论

学校读者我要写书评

暂无评论

Sign Language Recognition using Spiking Neural Networks 2

Sign Language Recognition using Spiking Neural Networks

引用

2nd International Conference on Machine Learning and Data engineering, ICMLDE 2023

作者： Chaudhari, Pranav Vicente-Sola, Alex Basu, Amlan Manna, Davide L. Kirkland, Paul Caterina, Gaetano Di Department of Information and Communication Technology Manipal Institute of Technology Manipal Academy of Higher Education Manipal576104 India Neuromorphic Sensor Signal Processing Lab Centre for Image and Signal Processing Electrical and Electronic Engineering Department University of Strathclyde Glasgow United Kingdom

In recent years, research in automatic Sign Language Recognition (SLR) has undergone significant progress, serving as a founda-tional base for developing applications that aim to promote the integration of deaf individuals into society. Most of this progress is owed to the recent developments in deep learning. However, the deployment of conventional Artificial Neural Networks (ANNs) can be hindered by their requirements in terms of computational power and energy consumption. Therefore, to improve the ef-ciency of current SLR systems, in this work, we propose the use of the increasingly popular Spiking Neural Networks (SNNs), which, on the one hand, provide more energy-efficient computations than conventional ANNs and, on the other hand, are able to process temporal sequences with simpler architectures thanks to their temporal dynamics. To evaluate our method, we utilize WLASL300, the 300-word (300 classes of signs) dataset from Word-Level American Sign Language, and achieve an improvement in accuracy with the SNN (+2.70%) over the previous state-of-the-art, when working with energy-efficient spiking neurons. Furthermore, we construct a non-spiking version of the same network and evaluate it in a similar manner. Our results demonstrate how the SNN has sparser activations (25% less), thanks to the use of spiking neurons, and therefore can be implemented with a lower power requirement than an ANN version of the same architecture. This work thus demonstrates the possibility of performing SLR in a very effective and efficient way, thus opening up the development of applications that span from the automatic real-time translation of dynamic signs to remote control utilizing sign languages. © 2024 Elsevier B.V.. All rights reserved.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Training Deep Neural Classifiers with Soft Diamond Regularizers

Training Deep Neural Classifiers with Soft Diamond Regulariz...

引用

International Conference on Machine Learning and Applications (ICMLA)

作者： Olaoluwa Adigun Bart Kosko Department of Electrical and Computer Engineering Signal and Image Processing Institute Los Angeles California

ISBN: (数字)9798350374889

ISBN: (纸本)9798350374896

We introduce new soft diamond regularizers that both improve synaptic sparsity and maintain classification accuracy in deep neural networks. These parametrized regularizers outperform the state-of-the-art hard-diamond Laplacian regularizer of Lasso regression and classification. They use thick-tailed symmetric alpha-stable $(\mathcal{S}\alpha \mathcal{S})$ bell-curve synaptic weight priors that are not Gaussian and so have thicker tails. The geometry of the diamond-shaped constraint set varies from a circle to a star depending on the tail thickness and dispersion of the prior probability density function. Training directly with these priors is computationally intensive because almost all $\mathcal{S}\alpha \mathcal{S}$ probability densities lack a closed form. A precomputed lookup table removed this computational bottleneck. We tested the new soft diamond regularizers with deep neural classifiers on the three datasets CIFAR-10, CIFAR-100, and Caltech-256. The regularizers improved the accuracy of the classifiers. The improvements included 4.57% on CIFAR-10, 4.27% on CIFAR-100, and 6.69% on Caltech-256. They also outperformed $L_{2}$ regularizers on all the test cases. Soft diamond regularizers also outperformed $L_{1}$ lasso or Laplace regularizers because they better increased sparsity while improving classification accuracy. Soft-diamond priors substantially improved accuracy on CIFAR-10 when combined with dropout, batch, or data-augmentation regularization.

关键词： Training Accuracy Sparse approximation Stars Tail Machine learning Probability density function Diamond Table lookup Residual neural networks

来源：评论

学校读者我要写书评

暂无评论

Segmentation of Retina Vessels in 2D OCT-Reconstructed Fundus images with 3D UNet 26

Segmentation of Retina Vessels in 2D OCT-Reconstructed Fundu...

引用

26th IEEE signal processing: Algorithms, Architectures, Arrangements, and Applications, SPA 2023

作者： Marciniak, Tomasz Stankiewicz, Agnieszka Zaradzki, Przemyslaw Poznan University of Technology Division of Electronic Systems and Signal Processing Institute of Automatic Control and Robotics Department of Control Robotics and Electrical Engineering Poznan Poland

ISBN: (纸本)9798350304985

This paper presents an analysis of the effectiveness of blood vessel segmentation in human retinal images. The segmentation process was carried out for three-dimensional optical coherence tomography (OCT) scans using the author's publicly available CAVRI-C database. A 3D version of the U-Net artificial neural network was used. Various dimensions of the input blocks were investigated to determine their influence on the segmentation process. The effectiveness of the proposed solutions was evaluated using various metrics and the results obtained were compared with the 2D solution, also based on neural networks. The use of the proposed solutions allowed us to improve the segmentation even by 15% (in the case of the F1-score). © 2023 Division of signal processing and Electronic systems, Poznan University of Technology (DSPES PUT).

关键词： Optical tomography

来源：评论

学校读者我要写书评

暂无评论

Controlled Causal Hallucinations Can Estimate Phantom Nodes in Multiexpert Mixtures of Fuzzy Cognitive Maps

arXiv

引用

arXiv 2024年

作者： Panda, Akash Kumar Kosko, Bart Signal and Image Processing Institute Department of Electrical and Computer Engineering University of Southern California United States

An adaptive multiexpert mixture of feedback causal models can approximate missing or phantom nodes in large-scale causal models. The result gives a scalable form of big knowledge. The mixed model approximates a sampled dynamical system by approximating its main limit-cycle equilibria. Each expert first draws a fuzzy cognitive map (FCM) with at least one missing causal node or variable. FCMs are directed signed partial-causality cyclic graphs. They mix naturally through convex combination to produce a new causal feedback FCM. Supervised learning helps each expert FCM estimate its phantom node by comparing the FCM's partial equilibrium with the complete multi-node equilibrium. Such phantom-node estimation allows partial control over these causal hallucinations and helps approximate the future trajectory of the dynamical system. But the approximation can be computationally heavy. Mixing the tuned expert FCMs gives a practical way to find several phantom nodes and thereby better approximate the feedback system's true equilibrium behavior. Copyright © 2024, The Authors. All rights reserved.

关键词： Phantoms

来源：评论

学校读者我要写书评

暂无评论

Bidirectional Backpropagation Autoencoding Networks for image Compression and Denoising

Bidirectional Backpropagation Autoencoding Networks for Imag...

引用

International Conference on Machine Learning and Applications (ICMLA)

作者： Olaoluwa Adigun Bart Kosko Department of Electrical and Computer Engineering Signal and Image Processing Institute Los Angeles California

关键词：

来源：评论

学校读者我要写书评

暂无评论

Training Deep Neural Classifiers with Soft Diamond Regularizers

arXiv

引用

arXiv 2024年

作者： Adigun, Olaoluwa Kosko, Bart Signal and Image Processing Institute Department of Electrical and Computer Engineering Los AngelesCA90089-2564 United States

We introduce new soft diamond regularizers that both improve synaptic sparsity and maintain classification accuracy in deep neural networks. These parametrized regularizers outperform the state-of-the-art hard-diamond Laplacian regularizer of Lasso regression and classification. They use thick-tailed symmetric alpha-stable (SαS) bell-curve synaptic weight priors that are not Gaussian and so have thicker tails. The geometry of the diamond-shaped constraint set varies from a circle to a star depending on the tail thickness and dispersion of the prior probability density function. Training directly with these priors is computationally intensive because almost all SαS probability densities lack a closed form. A precomputed look-up table removed this computational bottleneck. We tested the new soft diamond regularizers with deep neural classifiers on the three datasets CIFAR-10, CIFAR-100, and Caltech-256. The regularizers improved the accuracy of the classifiers. The improvements included 4.57% on CIFAR-10, 4.27% on CIFAR-100, and 6.69% on Caltech-256. They also outperformed L2 regularizers on all the test cases. Soft diamond regularizers also outperformed L1 lasso or Laplace regularizers because they better increased sparsity while improving classification accuracy. Soft-diamond priors substantially improved accuracy on CIFAR-10 when combined with dropout, batch, or data-augmentation regularization. Copyright © 2024, The Authors. All rights reserved.

关键词： Probability density function

来源：评论

学校读者我要写书评

暂无评论

Sign Language Recognition using Spiking Neural Networks

引用

Procedia Computer Science 2024年 235卷 2674-2683页

作者： Pranav Chaudhari Alex Vicente-Sola Amlan Basu Davide L. Manna Paul Kirkland Gaetano Di Caterina Department of Information and Communication Technology Manipal Institute of Technology Manipal Academy of Higher Education Manipal 576104 India Neuromorphic Sensor Signal Processing Lab Centre for Image and Signal Processing Electrical and Electronic Engineering Department University of Strathclyde Glasgow UK

关键词： Spiking Neural Networks American Sign Language Sign Language Recognition Deep Learning

来源：评论

学校读者我要写书评

暂无评论

Experimental implementation of chaos-based video encryption using visible light communication technology

引用

Applied Optics 2025年第9期64卷 C101-C118页

作者： Fotso, Sorel Bagio Nono Mbé, Jimmi Hervé Talla Takotue, Jean Ernest Takougang, Armand Cyrille Nzeukou Ndjiongue, Alain Richard Research Unit of Condensed Matter Electronics and Signal Processing Department of Physics University of Dschang P.O. Box 67 Dschang Cameroon Research Unit of Automation and Applied Computer Department of Electrical Engineering IUT-FV Bandjoun University of Dschang P.O. Box: 134 Bandjoun Cameroon Laboratory of Industrial Systems and Environmental Engineering Fotso Victor University Institute of Technology University of Dschang P.O. Box 134 Bandjoun Cameroon School of Electrical and Information Engineering Faculty of Engineering and the Built Environment University of the Witwatersrand Johannesburg South Africa

Simulations are commonly used to develop and evaluate video encryption algorithms. Although these approaches are useful for demonstrating theoretical feasibility and algorithm performance, they neglect practical challenges and real-world communication conditions. This oversight creates a critical gap in evaluating their effectiveness and security in practical applications. In this paper, we describe a novel, to the best of our knowledge, video encryption scheme that employs a modified chaotic Colpitts oscillator, a customized Josephus problem, and ribonucleic acid (RNA) operations, and is implemented in visible light communication (VLC) environment. The modified Colpitts oscillator reveals significant phenomena, the most prominent for this study being chaos. The encryption scheme involves block scrambling and pixel-level permutation employing a customized Josephus problem and a column and row shifting technique. Diffusion is then achieved through a custom Josephus problem-oriented RNA operation combined with basic RNA operations and cipher block chaining (CBC). Based on the results, we found that the customized Josephus problem-based permutation performed more efficiently than traditional methods, whereas the column and row shifting permutation was more robust. By eliminating key sequence coding and item-to-item operations, the proposed RNA operation reduces computational overhead. This video encryption scheme is proven to be effective and resilient at securing video data transmitted over the VLC channel, and it constitutes a significant advancement in areas such as surveillance, medical imaging, and military communication requiring robust secured video data. © 2025 Optica Publishing Group. All rights, including for text and data mining (TDM), Artificial Intelligence (AI) training, and similar technologies, are reserved.

关键词： Medical imaging

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：