Correcting Sample Selection Bias by Unlabeled Data
NIPS
Oral
KMM
Unsupervised Domain Adaptation with Distribution Matching Machines
AAAI 2018
Oral
3.特征学习方法及其扩展
主要是浅层优化算法
paper
来源
Novelty
代码复现
简称
Domain Adaptation via Transfer Component Analysis
IEEE TNNLS
Best
TCA
Transfer Feature Learning with Joint Distribution Adaptation
CVPR
Best
需要
JDA
Domain Invariant and Class Discriminative Feature Learning for Visual Domain Adapation
IEEE TIP
Best
需要
DICD
Transfer Joint Matching for Unsupervised Domain Adaptation
CVPR
Oral
TJM
Unsupervised Domain Adaptation With Label and Structural Consistency
IEEE TIP
Oral
Joint Geometrical and Statistical Alignment for Visual Domain Adaptation
CVPR 2018
Oral
JGSA
Adaptation Regularization: A General Framework for Transfer Learning
IEEE TKDE
Oral
4.纯深度学习网络结构研究
基础中的基础!推荐先看Stanford CS231n课程,百度搜索即可,可以快速了解深度学习
paper
来源
Novelty
代码复现
简称
|
|ImageNet Classification with Deep Convolutional Neural Networks |NIPS|Best|需要|AlexNet|
|Very Deep Convolutional Networks for Large-Scale Image Recognition ||Best|需要|VGG|
|Deep Residual Learning for Image Recognition |CVPR 2016 Best paper|Best|需要|ResNet|
|Densely Connected Convolutional Networks |CVPR 2017 Best paper|Best|需要|DenseNet|
|Squeeze-and-Excitation Networks |CVPR|Best||SENet|
|Batch Normalization Accelerating Deep Network Training by Reducing Internal Covariate Shift |ICML|Best ||BN|
|Deep Networks with Stochastic Depth |ECCV 2016 spotlight|Oral|||
|Going deeper with convolutions |NIPS|Oral||InceptionV1/GoogLeNet|
|Rethinking the Inception Architecture for Computer Vision |CVPR|Oral||InceptionV2/V3|
|Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning |CVPR|Oral||InceptionV4/Inception-ResNet|
|Aggregated Residual Transformations for Deep Neural Networks |CVPR|Oral||ResNext|
5.生成对抗网络(GAN)
paper
来源
Novelty
代码复现
简称
Generative Adversarial Nets
NIPS
Best
需要
GAN
Wasserstein GAN
Best
WGAN
Conditional Generative Adversarial Nets
Oral
CGAN
Least Squares Generative Adversarial Networks
ICCV
Oral
LSGAN
6.深度迁移学习
Deep Domain Adaptation,针对分类问题,研究重点!!
paper
来源
Novelty
代码复现
简称
How transferable are features in deep neural networks
NIPS 2014
Best
Learning Transferable Features with Deep Adaptation Networks
ICML 2015
Best
需要
DAN
Unsupervised Domain Adaptation by Backpropagation
ICML 2015
Best
需要
DANN/RevGrad
Maximum Classifier Discrepancy for Unsupervised Domain Adaptation
CVPR 2018
Best
需要
MCD
Unsupervised Domain Adaptation with Residual Transfer Networks