Methods > General

Skip Connection Blocks

Skip Connection Blocks are building blocks for neural networks that feature skip connections. These skip connections 'skip' some layers allowing gradients to better flow through the network. Below you will find a continuously updating list of skip connection blocks:

METHOD YEAR PAPERS
Residual Block
2015 1448
Bottleneck Residual Block
2015 1139
Dense Block
2016 268
Inverted Residual Block
2018 258
ResNeXt Block
2016 107
Non-Local Block
2017 67
Wide Residual Block
2016 36
CBHG
2017 36
ShuffleNet Block
2017 28
SRGAN Residual Block
2016 21
Reversible Residual Block
2017 12
Inception-ResNet-v2-B
2016 12
Inception-ResNet-v2-C
2016 12
DPN Block
2017 10
Ghost Bottleneck
2019 8
MelGAN Residual Block
2019 8
DV3 Convolution Block
2017 8
Res2Net Block
2019 7
Pyramidal Residual Unit
2016 7
Pyramidal Bottleneck Residual Unit
2000 7
FBNet Block
2018 7
Global Context Block
2019 6
Efficient Channel Attention
2019 5
Selective Kernel
2019 4
Big-Little Module
2018 4
One-Shot Aggregation
2019 4
Dilated Bottleneck Block
2018 4
Dilated Bottleneck with Projection Block
2018 4
Two-Way Dense Layer
2018 3
SqueezeNeXt Block
2018 3
ParaNet Convolution Block
2019 3
CSPResNeXt Block
2019 2
Elastic Dense Block
2018 2
GBlock
2019 2
DBlock
2019 2
Conditional DBlock
2019 2
DVD-GAN GBlock
2019 2
DVD-GAN DBlock
2019 2
NVAE Generative Residual Cell
2020 2
NVAE Encoder Residual Cell
2020 2
Residual SRM
2019 1
EESP
2018 1
Strided EESP
2018 1
OSA (identity mapping + eSE)
2019 1
Elastic ResNeXt Block
2018 1