Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination #47

Open
chullhwan-song opened this issue Aug 18, 2018 · 1 comment

Comments

@chullhwan-song
Copy link
Owner

https://arxiv.org/abs/1805.01978

@chullhwan-song
Copy link
Owner Author

what ?

  • unlabeled data만 이용 - Can we learn a good feature representation that captures apparent similarity
    among instances, instead of classes, ..? = unsupervised approach.
  • 기존의 방법 능가 - our method surpasses the state of the-art on ImageNet classification by a large margin.
  • 128차원으로..

구조

image

  • 언급햇듯이, 클래스 단위가 아니라 instance 즉, image 한장을 마치 클래스 단위처럼..
  • softmax 계산량은?
    • 실제 이런 문제는 word embeddings 힌트를 얻을 수 있음.
    • 그래서, 이전 연구에서보면, 크게 hierarchical softmax, noise-contrastive estimation(NCE), negative sampling 등이 있는데 여기서는, NCE 방법 고려함. 참고로, tf에서 따로 구현되어 있음.
    • NCE 가 Nonparametric Softmax
  • 소스 공개 : https://github.com/zhirongw/lemniscate.pytorch

실험

image

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant