About: PCA Based Kernel Initialization for Convolutional Neural Networks

Facets (new session)
Description
Metadata
Settings
- owl:sameAs
- Inference Rule:

About: PCA Based Kernel Initialization for Convolutional Neural Networks Goto Sponge NotDistinct Permalink

An Entity of Type : schema:ScholarlyArticle, within Data Space : covidontheweb.inria.fr associated with source document(s)

Attributes	Values
type	Academic Article research paper schema:ScholarlyArticle
isDefinedBy	Covid-on-the-Web dataset
has title	PCA Based Kernel Initialization for Convolutional Neural Networks
Creator	Peng, Wei Chen, Junfeng Hu, Yang Liu, Ke Pan, Hongyue Rong, Yuxi Wang, Yifeng Wu, Fangmin Xue, Xingsi
Source	PMC
abstract	The initialization of Convolutional Neural Networks (CNNs) is about providing reasonable initial values for the convolution kernels and the fully connected layers. In this paper, we proposed a convolution kernel initialization method based on the two-dimensional principal component analysis (2DPCA), in which a parametric equalization normalization method is used to adjust the scale between each neuron weight. After that the weight initial value can be adaptively adjusted according to different data samples. This method enables each neuron to fully back-propagate errors and accelerate network model training. Finally, a network model was built and experiments were performed using Tanh and ReLU activation functions. The experimental results verify the effectiveness of the proposed method through the distribution of histograms and the curve comparison diagrams of model training.
has issue date	2020-07-11(xsd:dateTime)
bibo:doi	10.1007/978-981-15-7205-0_7
has license	no-cc
sha1sum (hex)	1ad917049674d271404a6fa30f0fa939a652663b
schema:url	https://doi.org/10.1007/978-981-15-7205-0_7
resource representing a document's title	PCA Based Kernel Initialization for Convolutional Neural Networks
has PubMed Central identifier	PMC7351687
schema:publication	Data Mining and Big Data
resource representing a document's body	covid:1ad917049674d271404a6fa30f0fa939a652663b#body_text
is schema:about of	named entity 'kernels' named entity 'network model' named entity 'CNNs' named entity 'experiments' named entity 'values' named entity 'Convolutional Neural Networks' named entity 'scale' named entity 'reasonable' named entity 'MODEL TRAINING' named entity 'PROPOSED METHOD' named entity 'NEURON' named entity 'NETWORK MODEL' named entity 'BASED' named entity 'ADJUSTED' named entity 'WEIGHT' named entity 'PERFORMED' named entity 'VERIFY' named entity 'ERRORS' named entity 'INITIAL' named entity 'CONVOLUTION' covid:arg/1ad917049674d271404a6fa30f0fa939a652663b named entity 'activation' named entity 'principal component analysis' named entity 'convolution' named entity 'initial' named entity 'curve' named entity 'performed' named entity 'Convolutional Neural Networks' named entity 'neuron' named entity 'network model' named entity 'convolution kernel' named entity 'neuron' named entity 'principal component analysis' named entity 'activation functions' named entity 'model training' named entity 'PCA' named entity 'Convolutional Neural Networks' named entity 'covariance matrix' named entity 'input and output' named entity 'random variable' named entity 'activation function' named entity 'eigenvector' named entity 'Jacobian' named entity 'algorithm' named entity 'fully connected' named entity 'activation functions' named entity 'uniform distribution' named entity 'matrices' named entity 'feature vector' named entity 'variance' named entity 'eigenvector' named entity 'activation function' named entity 'randomization' named entity 'convolution kernel' named entity 'unsupervised feature learning' named entity 'Algorithm' named entity 'CPU' named entity 'convolutional layer' named entity 'activation function' named entity 'random variable' named entity 'histograms' named entity 'training data set' named entity 'convolutional neural network' named entity 'convolution kernel' named entity 'matrices' named entity 'PCA' named entity 'algorithm' named entity 'super resolution' named entity 'Algorithm' named entity 'rectifier' named entity 'standard deviation' named entity 'feature map' named entity 'convolution kernel' named entity 'convolutional layers' named entity 'normal distribution'

Faceted Search & Find service v1.13.91 as of Mar 24 2020

Alternative Linked Data Documents: Sponger | ODE Content Formats:

RDF

ODATA

Microdata

About

OpenLink Virtuoso version 07.20.3229 as of Jul 10 2020, on Linux (x86_64-pc-linux-gnu), Single-Server Edition (94 GB total memory)
Data on this page belongs to its respective rights holders.
Virtuoso Faceted Browser Copyright © 2009-2024 OpenLink Software