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

2024/09/25: GSNs : Generative Stochastic Networks #12

Open
animesh3008 opened this issue Sep 18, 2024 · 3 comments
Open

2024/09/25: GSNs : Generative Stochastic Networks #12

animesh3008 opened this issue Sep 18, 2024 · 3 comments
Assignees

Comments

@animesh3008
Copy link
Collaborator

animesh3008 commented Sep 18, 2024

Hey guys, I will be discussing Generative Stochastic Networks in the next journal club on 25th Sept 2024. More information follows soon :)
I will primarily be discussing this paper: https://academic.oup.com/imaiai/article/5/2/210/2363406

Here is the original paper which is shorter and condensed: https://proceedings.mlr.press/v32/bengio14.html

@animesh3008 animesh3008 self-assigned this Sep 24, 2024
@animesh3008
Copy link
Collaborator Author

Hi Everyone,

As you may have already noticed, this paper is very long and covers many important concepts, but it does not include much supplementary material like blog posts or videos making it tough to understand. It will not be possible for me to cover all aspects in one session.

Therefore, tomorrow I will present a general overview of the paper without going into detailed proofs. We will then discuss the best possible way to read this paper and continue the discussion in much more detail next week. The session might be relatively shorter than usual.

@animesh3008
Copy link
Collaborator Author

animesh3008 commented Sep 25, 2024

Attached is the presentation from today's discussion:
20240925_GenerativeStochasticNetworks.pdf

@moritzschaefer
Copy link
Collaborator

moritzschaefer commented Sep 25, 2024

Here is what we agreed upon for the upcoming sessions:

  • Next one @shemlem will talk about Neural ODEs (looking forward)
  • After that one, @adam will make a recap session
    • homework for that session: distribute the modern papers, for everyone to skim through them
      • then everyone can share the concepts that are lacking and that we should cover

It was also noted that these topics would probably be helpful to cover

  • MCMC
  • and potentially Boltzmann machines (less important though) might be beneficial
  • score matching (we should probably stick around the gradient-field and similiar models for abit then)

Thanks a lot Animesh!

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

No branches or pull requests

2 participants