Skip to content
New issue

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

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

Already on GitHub? # to your account

Please add NeurIPS 2019 paper #10

Closed
negrinho opened this issue Oct 19, 2019 · 3 comments
Closed

Please add NeurIPS 2019 paper #10

negrinho opened this issue Oct 19, 2019 · 3 comments

Comments

@negrinho
Copy link

Towards modular and programmable architecture search
Paper: https://arxiv.org/abs/1909.13404
Code: https://github.com/negrinho/deep_architect

Thank you

@D-X-Y
Copy link
Owner

D-X-Y commented Oct 20, 2019

Nice paper, added it. Would you mind to let me know the type of its searching algorithm?

@D-X-Y D-X-Y closed this as completed Oct 20, 2019
@negrinho
Copy link
Author

The paper is more focused on designing a language to encode search spaces than on search algorithms. The language gives you a well-defined API with which search algorithms can interact with search spaces. This means that the implementations of search algorithms and search spaces can be decoupled to some extent. The code link I shared contains the implementation of the language and the framework that we built around it. In conclusion, this work is not about a new search algorithm or search space, but about how the introduction of this language can enable modularity in architecture search implementations. Hopefully that clarifies it :).

@D-X-Y
Copy link
Owner

D-X-Y commented Oct 22, 2019

@negrinho Very clear! Thanks for your explanation. Nice work.

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

No branches or pull requests

2 participants