[pdeseminar] Mircea Petrache - Tomorrow tuesday 21 at 12:00 hs

[seminariosedp]

Dear colleagues and PDE friends,

We have the pleasure to announce the next session of our CMM PDE seminar 
with a talk given by Prof. Mircea Petrache (Puc) be held tomorrow 
Tuesday, November 21st at 12:00 hs (Santiago time) in the Sala John Von 
Neumann (seventh floor), Beauchef 851.

Title: Approximate and partial symmetries in Deep Learning

Abstract: The image of a rotated cat still represents a cat: while this 
simple rule seems obvious to a human, it is not obvious to neural 
networks, which separately "learn" each new rotation of the same image. 
This applies to different groups of symmetries for images, graphs, 
texts, and other types of data. Implementing "equivariant" neural 
networks that respect symmetries, reduces the number of learned 
parameters, and helps improve their generalization properties outside 
the training set. On the other hand, in networks that "identify too 
much", that is, where we impose too many symmetries, the error begins to 
increase, due to not respecting the data. In work with S. Trivedi, 
(NeurIPS 2023), we quantify this tradeoff, which allows to define the 
optimal amount of symmetries in learning models. I will give an 
introduction to classical learning theory bounds, and our extension of 
the ideas to the study of "partial/approximate equivariance". In 
passing, I'll describe some possible directions for working with partial 
symmetries in specific tasks.

The link to access the zoom meeting: https://reuna.zoom.us/j/5602423930 
[1]

You can find more information about the seminar and future sesions in 
the seminar web page:
https://eventos.cmm.uchile.cl/pdeseminar/ [2]

Best wishes,

Rayssa Caju and María Eugenia Martínez.

Links:
------
[1] https://reuna.zoom.us/j/5602423930
[2] https://eventos.cmm.uchile.cl/pdeseminar/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listas.dim.uchile.cl/pipermail/pdeseminar/attachments/20231120/8d4d6fbe/attachment.html>