Here are the final slides in case anyone wants them :)
https://docs.google.com/presentation/d/1yqVPqjplEcxSRrqRAIaQtGJR5Z94e0-kN9Ll...
PS: If you have any questions send me an email or stop me on the halls! (I'm here till Thursday next week).
Best! Martin
2017-03-07 14:09 GMT-05:00 Dzmitry Bahdanau dimabgv@gmail.com:
This is a reminder about the additional tea-talk, which is going to happen in 20 minutes.
Here is also a *short bio* from Martin:
I'm Martin Arjovsky, I'm currently doing my PhD at New York University, and being advised by Léon Bottou. I did my undergraduate and master's in the University of Buenos Aires, Argentina (my home country). In the middle I took a year off to do internships in different places (Google, Facebook, Microsoft, and the Université de Montréal). My master's thesis advisor was Yoshua Bengio, who also advised me during my stay at UdeM. In general I'm interested in the intersection between learning and mathematics, how we can ground the different learning processes that are involved in different problems, and leverage this knowledge to develop better algorithms.
On Sun, 5 Mar 2017 at 21:58 Dzmitry Bahdanau dimabgv@gmail.com wrote:
Hi all,
This Tuesday (March 7) we will have an extra tea-talk by *Martin Arjovsky*. Please note that the time is different (14:30), but the place is the same (AA6214). Hope to see many of you there!
*Title: *On Different Distances Between Distributions and Generative Adversarial Networks
*Abstract: *Generative adversarial networks (GANs) are notoriously difficult to train. At the core of it, we show that these problems arise naturally when trying to learn distributions whose support lie in low dimensional manifolds. We show how these problems are consequences of trying to optimize the classical divergences (KL, JSD, etc) between our real and data distribution, and that these are symptoms of a more general phenomenon, pointing towards the inefficacy of the usual divergences in certain settings. After that, we bring into play the Wasserstein distance, which we prove doesn't suffer from the same behaviour, and provide a first step on an algorithm that tries to approximately optimize this distance.
Dima
Lisa_labo mailing list Lisa_labo@iro.umontreal.ca https://webmail.iro.umontreal.ca/mailman/listinfo/lisa_labo