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

[34] VQ-VAE: Neural Discrete Representation Learning #34

Open
Dongwoo-Im opened this issue May 8, 2023 · 1 comment
Open

[34] VQ-VAE: Neural Discrete Representation Learning #34

Dongwoo-Im opened this issue May 8, 2023 · 1 comment
Assignees

Comments

@Dongwoo-Im
Copy link
Contributor

Links

한 줄 요약

  • VAE의 continuous latent variable 대신 discrete latent variable을 사용하는 VQ-VAE 제안
  • Discretisation 과정에 vector quantisation (VQ)을 적용함으로써, VAE의 posterior collapse 현상을 방지할 수 있다고 주장
    • VQ-VAE가 latent discretisation을 처음으로 제안한 것은 아니지만 VAE와 비슷한 성능을 보였다는 점에서 유의미함
    • Posterior collapse는 approximate posterior가 gaussian prior z와 같아지는 현상으로 확인됨 (정확한 표현은 아닐 수 있음)

선택 이유

  • 유명한 논문이어서 + DALL-E를 읽기 위해 선정
@Dongwoo-Im
Copy link
Contributor Author

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