Representational Power of Graph Neural Networks - Stefanie Jegelka

Workshop on Theory of Deep Learning: Where next?
Topic: Representational Power of Graph Neural Networks
Speaker: Stefanie Jegelka
Affiliation: Massachusetts Institute of Technology
Date: October 18, 2019
For more video please visit video.ias.edu

[@jackflash9123]

Nice, really clear and informative talk! Thanks for the upload :)

[@jackflash9123]

@Casey Dallas Don't know if this is the right place to promote your scam.

[@normanhofer8965]

Thanks, very informative!

[@taku8751]

Is that better than transformer?

[@Onislayer]

Sum is not injective...

[@wandbfanatic92]

True, but still better than mean, which is better than max.

[@martinschulze5399]

why is a sum not injektive o.0?

[@Bridgelessalex]

​@@martinschulze53992+2 = 1+3

[@jayt696969]

Fascinating topic. But professor, you speak a bit too fast