Learning Deep and Wide: A Spectral Method for Learning Deep Networks

Shao L.; Wu D.; Li X.

首页> 外文期刊>Neural Networks and Learning Systems, IEEE Transactions on >Learning Deep and Wide: A Spectral Method for Learning Deep Networks

【24h】

Learning Deep and Wide: A Spectral Method for Learning Deep Networks

机译：深入学习：学习深度网络的一种光谱方法

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相似文献
相关主题

摘要

Building intelligent systems that are capable of extracting high-level representations from high-dimensional sensory data lies at the core of solving many computer vision-related tasks. We propose the multispectral neural networks (MSNN) to learn features from multicolumn deep neural networks and embed the penultimate hierarchical discriminative manifolds into a compact representation. The low-dimensional embedding explores the complementary property of different views wherein the distribution of each view is sufficiently smooth and hence achieves robustness, given few labeled training data. Our experiments show that spectrally embedding several deep neural networks can explore the optimum output from the multicolumn networks and consistently decrease the error rate compared with a single deep network.

机译：构建能够从高维感官数据中提取高级表示的智能系统，是解决许多与计算机视觉相关的任务的核心。我们提出多光谱神经网络（MSNN），以从多列深度神经网络中学习特征，并将倒数第二个分层判别流形嵌入到紧凑表示中。低维嵌入探索了不同视图的互补特性，其中在给定标记训练数据很少的情况下，每个视图的分布足够平滑，因此可以实现鲁棒性。我们的实验表明，频谱嵌入多个深度神经网络可以探索多列网络的最佳输出，并且与单个深度网络相比，始终可以降低错误率。

著录项

来源
《Neural Networks and Learning Systems, IEEE Transactions on》 |2014年第12期|2303-2308|共6页
作者
Shao L.; Wu D.; Li X.;
展开▼
作者单位

College of Electronic and Information Engineering, Nanjing University of Information Science and Technology, Nanjing, China;

展开▼
收录信息
原文格式 PDF
正文语种 eng
中图分类
关键词
Data models; Error analysis; Feature extraction; Laplace equations; Neural networks; Noise; Training; Deep networks; multispectral embedding; representation learning; representation learning.;

机译：数据模型;误差分析;特征提取;拉普拉斯方程;神经网络;噪声;训练;深度网络;多光谱嵌入;表示学习;表示学习;

相似文献

外文文献
中文文献
专利

1. Active Transfer Learning Network: A Unified Deep Joint Spectral–Spatial Feature Learning Model for Hyperspectral Image Classification [J] . Deng Cheng, Xue Yumeng, Liu Xianglong, IEEE Transactions on Geoscience and Remote Sensing . 2019,第3期

机译：主动转移学习网络：用于高光谱图像分类的统一的深度联合光谱空间特征学习模型
2. A deep learning method for bearing fault diagnosis based on Cyclic Spectral Coherence and Convolutional Neural Networks [J] . Zhuyun Chen, Alexandre Mauricio, Weihua Li, Mechanical systems and signal processing . 2020,第Juna期

机译：基于循环谱相干和卷积神经网络的轴承故障诊断深度学习方法
3. Deep-learning: investigating deep neural networks hyper-parameters and comparison of performance to shallow methods for modeling bioactivity data [J] . Alexios Koutsoukas, Keith J. Monaghan, Xiaoli Li, Journal of Cheminformatics . 2017,第1期

机译：深度学习：研究深度神经网络的超参数并将性能与浅层方法进行生物活性数据建模的比较
4. Deep Learning Angiography (DLA): Three-dimensional C-arm cone beam CT angiography generated from deep learning method using a convolutional neural network [C] . Juan C. Montoya, Yinsheng Li, Charles Strother, SPIE Medical Imaging Conference . 2018

机译：深度学习血管造影（DLA）：使用卷积神经网络从深度学习方法生成的三维C型臂锥束CT血管造影
5. Plant Segmentation by Supervised Machine Learning Methods and Phenotypic Trait Extraction of Soybean Plants Using Deep Convolutional Neural Networks with Transfer Learning [D] . Adams, Jason R. 2020

机译：植物分割通过深度卷积神经网络与转移学习的豆豆植物的植物分割和表型特性
6. Deep-learning: investigating deep neural networks hyper-parameters and comparison of performance to shallow methods for modeling bioactivity data [O] . Alexios Koutsoukas, Keith J. Monaghan, Xiaoli Li, 2017

机译：深度学习：研究深度神经网络的超参数并将性能与浅层方法进行生物活性数据建模的比较
7. Active Transfer Learning Network: A Unified Deep Joint Spectral–Spatial Feature Learning Model for Hyperspectral Image Classification [O] . Cheng Deng, Yumeng Xue, Xianglong Liu, 2019

机译：主动转移学习网络：高光谱图像分类的统一深度联接光谱空间特征学习模型

Learning Deep and Wide: A Spectral Method for Learning Deep Networks

摘要

著录项

相似文献

相关主题

期刊订阅