Date de tri

DATAIA Seminar | « Solving inverse problems with invertible neural networks » - Ullrich Köthe

Bandeau image
Ulrich-Seminar

DATAIA Seminar | « Solving inverse problems with invertible neural networks » - Ullrich Köthe

  • Practical information
Date limite d'inscription
Closing date for registration : 24.09.20
Lieu de l'événement
URL : https://bluejeans.com/9352872428/9913
Date de l'événement (intitulé)
24 september 2020 - 15.00
Chapo
As part of its scientific activities, the DATAIA Institute organises monthly seminars aimed at discussing about AI.
Contenu
Corps de texte

Ullrich Köthe (Université de Heidelberg) is leading the DATAIA seminar of September on « Solving inverse problems with invertible neural networks ».

Interpretable models are a hot topic in neural network research. This talk will focus on inverse problems, where one wants to infer backwards from observations to the hidden characteristics of a system. I will focus on three aspects of interpretability: reliable uncertainty quantification, outlier detection, and disentanglement into meaningful features. It turns out that invertible neural networks -- networks that work equally well in the forward and inverse direction -- are great tools for that kind of analysis: They act as non-linear generalizations of classical methods like PCA and ICA. Examples from physics, medicine, and computer vision demonstrate the practical utility of the new method.

Ancre
Practical information
Corps de texte

The webinar will take place on September 24th 2020 at 15.00 and it will be live broadcasted.

It is recommended to use Google Chrome, Firefox, or the BlueJeans app (https://www.bluejeans.com/downloads) to join the webinar.

 

Don't miss out any new announcement about a new DATAIA seminar!
Subscribe to the DATAIA seminars mailing-list by clicking here

Registration form
Would you like to subscribe to our seminars' mailing-list?