数据标注

 MICCAI 

【回顾】视见医疗陈浩：从MICCAI2017一窥医疗影像的最近进展

multi instance learning

active learning

Fine-tuning Convolutional Neural Networks for Biomedical Image Analysis： Actively and Incrementally

Active learning-Fine-tuning Convolutional Neural Networks for Biomedical Image Analysis: Actively a

【人工智能】Active Learning: 一个降低深度学习时间，空间，经济成本的解决方案

Suggestive Annotation: A Deep Active Learning Framework for Biomedical Image Segmentation

dual learning

对偶学习：一种新的机器学习范式，数据标注成本从2000万美元降到200万

https://github.com/NonameAuPlatal/Dual_Learning

https://github.com/thompsonb/DL4MT

伪标签学习

【译文】伪标签学习导论 - 一种半监督学习方法

https://github.com/shubhamjn1/Pseudo-Labelling---A-Semi-supervised-learning-technique

伪标签：教你玩转无标签数据的半监督学习方法

Pseudo-Label : The Simple and Efficient Semi-Supervised Learning Method for Deep Neural Networks

弱监督object detection

Weakly Supervised Deep Detection Networks

https://github.com/zhusiwei93/Weakly-Supervised-Object-Detection

https://github.com/yanxp/Weakly-object-detection

弱监督分割

见微知著：语义分割中的弱监督学习

论文笔记 | BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentation

【图像】初识弱监督下的图像语义分割

融合先验知识

deep learning with Domain Knowledge

Label-Free Supervision of Neural Networks with Physics and Domain Knowledge

https://www.leiphone.com/news/201702/VwOQays57XUF0aT5.html

http://www.sohu.com/a/125922514_465975

Automatic Landmark Estimation for Adolescent Idiopathic Scoliosis Assessment Using BoostNet

Integrating statistical prior knowledge into convolutional neural network

深度学习（机器学习）的下一步如何发展？