-
In the original paper CapsuleGAN: Generative Adversarial Capsule Network, they have used semi supervised dataset.How do we incorporate that data in this model??
-
write the outlines in the following format:
# Introduction
* introduce classification task with different data settings ordinary setting, weak supervision, noisy label, and complementary label l…
-
MIT-SelefDriving Course 笔记
--------------------------------------
1. Difference between ML(Machine Learning) & DL (Deep Learning)
- ML: input -> Feature extractoiin (human beings to do that) -> …
Iranb updated
5 years ago
-
# Abstract
Generative Adversarial Networks(GAN)은 데이터 생성에서 뛰어난 모습을 보이고 있다. 많은 영역에서 쓰이고 있지만 여전히 안정적인 학습에는 어려움이 따른다. 문제점으로는 Nash-equilibrium, internal covariate shift, mode collapse, vanishing gradient,…
-
**Proceedings**
https://papers.nips.cc/book/advances-in-neural-information-processing-systems-30-2017
https://github.com/catpanda/NIPS_2017
**PaperLists (#Papers 679)**
https://www.dropbox.com/s…
-
Here are a few things I found related to the use of the generative models to obtain synthetic medical data.
## Bibliography
- **Synthesis in Multi-Contrast MRI with Conditional Generative Adver…
-
## 内容
用mixup为对比学习做数据增强，可以在SimCLR，MoCo，BYOL上用，在图像、语音和表格数据（为啥这类文章都不做文本？）上都有提升。也验证了方法的正则化效果，在（1）数据不够（2）数据增强的领域知识不够，两种情况下明显改善了对比学习。
具体就是为batch中每个instance分配virtual label（其实就是它的序号对应的one-hot），然后再mixup生成新…
-
-
自然语言处理（Natural Language Processing, NLP）非常有意思！互联网上最不缺乏的就是文本数据了，如何处理这些数据来获取有价值的信息，这是一个非常值得探索的领域。
在这个Issues下面，我主要记录一些阅读过还不错的文献，做点摘记。
-
190126~