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

Question for your paper MSD #5

Open
curryandsun opened this issue Nov 11, 2020 · 3 comments
Open

Question for your paper MSD #5

curryandsun opened this issue Nov 11, 2020 · 3 comments

Comments

@curryandsun
Copy link

Hi,
Thanks for sharing this code and it's really helpful.

Recently I read your paper:"MSD: Multi-Self-Distillation Learning via Multi-classifiers within Deep Neural Networks".It's a very interesting work and the results are much better than the paper "be your own teacher" which you reimplement here.

However,after reading your paper,I could just find some slight differences between this two papers:
1.the differences of bottleneck in the model.
2.some changes of hyper-parameter.

Is there some important details that I missed?And could you please tell me about the key difference between the two papers that lead to such a significant improvement?

@luanyunteng
Copy link
Owner

MSD is a bi-KD or mutual learning framework while the 'be your own teacher' is a one-way knowledge distillation method

@luanyunteng
Copy link
Owner

Hi,
Thanks for sharing this code and it's really helpful.

Recently I read your paper:"MSD: Multi-Self-Distillation Learning via Multi-classifiers within Deep Neural Networks".It's a very interesting work and the results are much better than the paper "be your own teacher" which you reimplement here.

However,after reading your paper,I could just find some slight differences between this two papers:
1.the differences of bottleneck in the model.
2.some changes of hyper-parameter.

Is there some important details that I missed?And could you please tell me about the key difference between the two papers that lead to such a significant improvement?

MSD is a bi-KD or mutual learning framework while the 'be your own teacher' is a one-way knowledge distillation method

@yoona-unicorn
Copy link

hi,
Recently I read the paper "MSD" and can you provide code about this paper? thanks

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

3 participants