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

Meeting with Aline and Carol about IVA #21

Open
Kotzly opened this issue Dec 23, 2022 · 2 comments
Open

Meeting with Aline and Carol about IVA #21

Kotzly opened this issue Dec 23, 2022 · 2 comments

Comments

@Kotzly
Copy link
Owner

Kotzly commented Dec 23, 2022

The goal of this issue is to dump the meeting notes with Aline and Carol. The former has participated in my Master's qualification exam, and the later is PhD student that has worked with Independent Vector Analysis.

Meeting minutes:

  • Question in which algorithm has been implemented/tested: Matrix natural gradient descent, Vector gradient descent, Newton SCV Update (row-wise), Block-Newton SCV update.
  • Multivariate Gaussian vs Multivariate Laplace distribution of sources
  • Question about Inter subject and extension to unseen subjects
  • Normalization of unseen subjects
  • IVA in Motor Imagery
@Kotzly
Copy link
Owner Author

Kotzly commented Dec 23, 2022

  • Qual o benchmark?
  • Um IVA por classe, na inferencia passa por todos IVA
  • Vai me passar os codigos da tulay pra IVA
  • Codigo dos IVAs
  • Possibilidades de Inter-subject: Encontrar qual matriz de treino tem mais relacao com um sujeito de teste; usar um pedaço do teste.
  • Usar direçoes do branqueamento pra discriminação de classes
  • 66% acc em 2 classes

@Kotzly
Copy link
Owner Author

Kotzly commented Dec 24, 2022

Bom dia Paulo, Romis e Sara,

Obrigada pelas sugestões e por organizarem a reunião de hoje, a conversa foi muito produtiva e espero revê-los em 2023.

Segue anexo o artigo com relação ao benchmark do dataset 2a. Além disso, também segue abaixo os links de acesso aos códigos do IVA em Matlab e ao código em Python desenvolvido por uma aluna da Tullay:

Link de acesso aos códigos do IVA em Matlab:
http://mlsp.umbc.edu/resources.html#Multiset

Link de acesso ao código do IVA em Python:
https://github.com/SSTGroup/independent_vector_analysis

Qualquer dúvida, estou à disposição.

Boas Festas!

Atenciosamente,
Caroline.

Unsupervised_feature_extraction_with_autoencoders_for_EEG_based_multiclass_motor_imagery_BCI.pdf

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

1 participant