1998
LeNet-5 卷积神经网络
- 《Gradient-based learning applied to document recognition》
- http://yann.lecun.com/exdb/publis/pdf/lecun-01a.pdf
2012
AlexNet 卷积神经网络
- 《ImageNet Classification with Deep Convolutional Neural Networks》
- https://www.cs.toronto.edu/~fritz/absps/imagenet.pdf
2013
卷积神经网络可视化
- 《Visualizing and Understanding Convolutional Networks》
- https://arxiv.org/pdf/1311.2901.pdf
Network In Network结构
- 《Network In Network》
- https://arxiv.org/pdf/1312.4400.pdf
2014
GAN 神经网络
- 《Generative Adversarial Network》
- https://arxiv.org/pdf/1406.2661.pdf
VGG 卷积神经网络
- 《Very Deep Convolutional Networks for Large-scale Image Recognition》
- https://arxiv.org/pdf/1409.1556.pdf
Inception模组
- 《Going deeper with convolutions》
- https://arxiv.org/pdf/1409.4842.pdf
全卷积网络
- 《Fully Convolutional Networks for Semantic Segmentation》
- https://arxiv.org/pdf/1411.4038.pdf
纯卷积网络
- 《Striving for Simplicity: The All Convolutional Net》
- https://arxiv.org/pdf/1412.6806.pdf
2015
生成短句
- 《Neural Responding Machine for Short-Text Conversation》
- https://arxiv.org/pdf/1503.02364.pdf
具有一定体系结构约束的GAN
- 《Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks》
- https://arxiv.org/pdf/1511.06434.pdf
Inception 模组的衍生
- 《Rethinking the Inception Architecture for Computer Vision》
- https://arxiv.org/pdf/1512.00567.pdf
残差网络
- 《Deep Residual Learning for Image Recognition》
- https://arxiv.org/pdf/1512.03385.pdf
2016
Inception 第4版
- 《Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning》
- https://arxiv.org/pdf/1602.07261.pdf
SqueezeNet
- 《Alexnet-level Accuracy with 50x Fewer Parameters and <0.5MB Model Size》
- https://arxiv.org/pdf/1602.07360.pdf
排列对残差网络的影响
- 《Identity Mappings in Deep Residual Networks》
- https://arxiv.org/pdf/1603.05027.pdf
随机深度
- 《Deep Networks with Stochastic Depth》
- https://arxiv.org/pdf/1603.09382.pdf
从网络集合的角度理解残差网络
- 《Residual Networks Behave Like Ensembles of Relatively Shallow Networks》
- https://arxiv.org/pdf/1605.06431.pdf
广度残差网络
- 《Wide Residual Networks》
- https://arxiv.org/pdf/1605.07146.pdf
FractalNet
- 《Ultra-deep Neural Networks without Residuals》
- https://arxiv.org/pdf/1605.07648.pdf
Info GAN 网络 - 具有特征的对抗生成网络
- 《InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets》
- https://arxiv.org/pdf/1606.03657.pdf
DenseNet
- 《Densely Connected Convolutional Networks》
- https://arxiv.org/pdf/1608.06993.pdf
SR GAN 网络 - 实现超分辨率
- 《Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network》
- https://arxiv.org/pdf/1609.04802.pdf
Xception 网络架构
- 《Xception: Deep Learning with Depthwise Separable Convolutions》
- https://arxiv.org/pdf/1610.02357.pdf
Pyramid Net
- 《Deep Pyramidal Residual Networks》
- https://arxiv.org/pdf/1610.02915.pdf
将深度学习用于识别罪犯
- 《Responses to Critiques on Machine Learning of Criminality Perceptions》
- https://arxiv.org/pdf/1611.04135.pdf
ResNext
- 《Aggregated Residual Transformations for Deep Neural Networks》
- https://arxiv.org/pdf/1611.05431.pdf
GAN 网络的应用
- 《Plug & Play Generative Networks》
- https://arxiv.org/pdf/1612.00005.pdf
用句子描述图片
- 《Knowing When to Look: Adaptive Attention via A Visual Sentinel for Image Captioning》
- https://arxiv.org/pdf/1612.01887.pdf
2017
Wassertein GAN 神经网络
- 《Wassertein GAN》
- https://arxiv.org/pdf/1701.07875.pdf
从梯度的角度理解残差网络
- 《The Shattered Gradients Problem》
- https://arxiv.org/pdf/1702.08591.pdf
示例分割
- 《Mask R-CNN》
- https://arxiv.org/pdf/1703.06870.pdf
Cycle GAN 神经网络
- 《Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks》
- https://arxiv.org/pdf/1703.10593.pdf
Shake-Shake正则化
- 《Shake-Shake regularization》
- https://arxiv.org/pdf/1705.07485.pdf
Dual Path Net
- 《Dual Path Networks》
- https://arxiv.org/pdf/1707.01629.pdf
3亿个样本,令模型效果提升
- 《Revisiting Unreasonable Effectiveness of Data in Deep Learning Era》
- https://arxiv.org/pdf/1707.02968.pdf