|
gptkbp:instance_of
|
gptkb:television_channel
|
|
gptkbp:bfsLayer
|
3
|
|
gptkbp:bfsParent
|
gptkb:Yann_Le_Cun
|
|
gptkbp:applies_to
|
Facial recognition
Image classification
Object detection
|
|
gptkbp:architectural_style
|
Feedforward neural network
|
|
gptkbp:consists_of
|
Convolutional layers
Fully connected layers
Pooling layers
|
|
gptkbp:developed_by
|
gptkb:Yann_Le_Cun
|
|
gptkbp:has_method
|
Average pooling
Batch size
Learning rate
Number of epochs
Subsampling
|
|
gptkbp:has_sequel
|
Subsampling layer
|
|
gptkbp:has_skin
|
Output layer
Convolutional layer
Fully connected layer
Subsampling layer
|
|
gptkbp:has_transformation
|
Sigmoid
Tanh
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Le Net architecture
|
|
gptkbp:influenced_by
|
Backpropagation algorithm
|
|
gptkbp:input_output
|
32x32 pixels
|
|
gptkbp:is_considered_as
|
Pioneering work in CN Ns
|
|
gptkbp:is_implemented_in
|
gptkb:Graphics_Processing_Unit
gptkb:Keras
gptkb:Py_Torch
|
|
gptkbp:is_popular_in
|
Deep learning community
|
|
gptkbp:is_related_to
|
gptkb:Artificial_Intelligence
Machine learning
Computer vision
|
|
gptkbp:is_used_for
|
Handwritten digit recognition
|
|
gptkbp:is_used_in
|
Research papers
Industry applications
|
|
gptkbp:losses
|
Mean squared error
Cross-entropy loss
|
|
gptkbp:number_of_cores
|
7 layers
|
|
gptkbp:performance
|
gptkb:Image_Net
gptkb:CIFAR-10
gptkb:SVHN
gptkb:Fashion_MNIST
gptkb:MNIST
|
|
gptkbp:predecessor
|
gptkb:Neocognitron
|
|
gptkbp:resolution
|
10 classes
|
|
gptkbp:successor
|
gptkb:Alex_Net
|
|
gptkbp:training
|
Stochastic gradient descent
Mini-batch gradient descent
|
|
gptkbp:year_created
|
gptkb:1989
|