常见神经网络出处及下载链接
- 一、CNN及其变体
- 二、RNN及其变体
- 三、Transformer及其变体
- 四、生成对抗网络GAN及其变体
- 五、特殊网络
- 总览
整理下这些年主要的神经网络文献,包括了三大特征提取器(CNN/RNN/Transformer)、GAN等一些混合模型。不定期补充更新
一、CNN及其变体
序号 |
网络结构 |
文献出处 |
年份 |
1 |
LeNet-5(CNN起源) |
Generalization and Network Design Strategies |
1989 |
2 |
VGGNet |
VERY DEEP CONVOLUTIONAL NETWORKS FOR LARGE-SCALE IMAGE RECOGNITION |
2014 |
3 |
Inception |
Rethinking the Inception Architecture for Computer Vision |
2016 |
4 |
ResNet |
Deep Residual Learning for Image Recognition |
2016 |
5 |
DenseNet |
Densely Connected Convolutional Networks |
2017 |
二、RNN及其变体
序号 |
网络结构 |
文献出处 |
年份 |
1 |
RNN(起源) |
Learning representations by back-propagating errors |
1986 |
2 |
LSTM |
Long short-term memory |
1997 |
3 |
BiLSTM |
Framewise phoneme classification with bidirectional LSTM and other neural network architectures |
2005 |
4 |
GRU |
Learning Phrase Representations Using RNN Encoder–Decoder for Statistical Machine Translation |
2014 |
三、Transformer及其变体
序号 |
网络结构 |
文献出处 |
年份 |
1 |
Transformer(起源) |
Attention is all you need |
2017 |
2 |
Vision Transformer (ViT) |
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale |
2020 |
3 |
Pooling-based Vision Transformer (PiT) |
Rethinking spatial dimensions of vision transformers |
2021 |
四、生成对抗网络GAN及其变体
序号 |
网络结构 |
文献出处 |
年份 |
1 |
GAN(起源) |
Generative Adversarial Networks |
2014 |
2 |
CGAN |
Conditional Generative Adversarial Nets |
2014 |
3 |
InfoGAN |
Infogan: Interpretable representation learning by information maximizing generative adversarial nets |
2016 |
五、特殊网络
这里包括了混合模型以及特殊的图卷积网络GCN,虽然GCN借鉴了CNN的思想,但是从结构上和计算方式上,差异还是挺大的。所以单独在这里列出。
序号 |
网络结构 |
文献出处 |
年份 |
1 |
CLDNN |
Convolutional, Long Short-Term Memory, fully connected Deep Neural Networks |
2015 |
2 |
GCN(图卷积) |
Semi-supervised Classification with Graph Convolutional Networks |
2016 |
总览
序号 |
网络结构 |
文献出处 |
年份 |
1 |
RNN(序列模型起源) |
Learning representations by back-propagating errors |
1986 |
2 |
CNN(卷积网络起源) |
Generalization and Network Design Strategies |
1989 |
3 |
LSTM |
Long short-term memory |
1997 |
4 |
BiLSTM |
Framewise phoneme classification with bidirectional LSTM and other neural network architectures |
2005 |
5 |
VGGNet |
VERY DEEP CONVOLUTIONAL NETWORKS FOR LARGE-SCALE IMAGE RECOGNITION |
2014 |
6 |
GRU |
Learning Phrase Representations Using RNN Encoder–Decoder for Statistical Machine Translation |
2014 |
7 |
GAN |
Generative Adversarial Networks |
2014 |
8 |
CGAN |
Conditional Generative Adversarial Nets |
2014 |
9 |
CLDNN |
Convolutional, Long Short-Term Memory, fully connected Deep Neural Networks |
2015 |
10 |
Inception |
Rethinking the Inception Architecture for Computer Vision |
2016 |
11 |
ResNet |
Deep Residual Learning for Image Recognition |
2016 |
12 |
GCN(图卷积) |
Semi-supervised Classification with Graph Convolutional Networks |
2016 |
13 |
InfoGAN |
Infogan: Interpretable representation learning by information maximizing generative adversarial nets |
2016 |
14 |
DenseNet |
Densely Connected Convolutional Networks |
2017 |
15 |
Transformer(起源) |
Attention is all you need |
2017 |
16 |
Vision Transformer (ViT) |
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale |
2020 |
17 |
Pooling-based Vision Transformer (PiT) |
Rethinking spatial dimensions of vision transformers |
2021 |
情感课堂