|
gptkbp:instanceOf
|
gptkb:Neural_Network_Component
|
|
gptkbp:architecture
|
gptkb:Skip_Connection
Identity Mapping
|
|
gptkbp:citation
|
100000+
|
|
gptkbp:commonActivation
|
gptkb:ELU
gptkb:ReLU
gptkb:Leaky_ReLU
|
|
gptkbp:component
|
gptkb:ResNet-101
gptkb:ResNet-152
gptkb:ResNet-18
gptkb:ResNet-34
gptkb:ResNet-50
|
|
gptkbp:enables
|
Training of very deep networks
|
|
gptkbp:hasVariant
|
Bottleneck Residual Block
Dilated Residual Block
Pre-activation Residual Block
Squeeze-and-Excitation Residual Block
|
|
gptkbp:influenced
|
gptkb:transformation
gptkb:Highway_Network
gptkb:UNet
Dense Block
|
|
gptkbp:input
|
Input Tensor
|
|
gptkbp:introduced
|
gptkb:Shaoqing_Ren
gptkb:Xiangyu_Zhang
gptkb:Jian_Sun
gptkb:Kaiming_He
|
|
gptkbp:introducedIn
|
gptkb:ResNet
2015
|
|
gptkbp:notablePublication
|
gptkb:Deep_Residual_Learning_for_Image_Recognition
https://arxiv.org/abs/1512.03385
|
|
gptkbp:operator
|
Convolution
Addition
Batch Normalization
Activation Function
|
|
gptkbp:output
|
Output Tensor
|
|
gptkbp:purpose
|
Mitigate vanishing gradient problem
|
|
gptkbp:usedIn
|
gptkb:Natural_Language_Processing
gptkb:Speech_Recognition
Deep Learning
Object Detection
Image Classification
Semantic Segmentation
|
|
gptkbp:bfsParent
|
gptkb:Residual_Network
|
|
gptkbp:bfsLayer
|
8
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Residual Block
|