Author: kyunghyuncho

I recently give one lecture on how to use neural nets for natural language processing at the course <INF8225 – Intelligence artificielle : techniques probabilistes et d’apprentissage> at the École Polytechnique de Montréal. It was a very fun experience having extremely lively discussion with active students.  The slides can be found at https://drive.google.com/file/d/0B16RwCMQqrtdNEhwbHN2bXJzdXM/view?usp=sharing.

I will co-organize the Deep Learning Workshop at ICML 2015 in Lille, France together with Geoff Hinton, Yann LeCun, Yoshua Bengio, Max Welling and Durk Kingma. This is my first time organizing a workshop (or as a matter of fact, any large academic meeting), and I have made a few blunders so far, but I believe this workshop will be pretty awesome with an awesome lineup of invited speakers:   First day (July 10) Tara Sainath, Google Yann Ollivier, Paris-Sud University Oriol Vinyals, Google Jason Weston, Facebook Jorge Nocedal, Northwestern University   Second day (July 11) Neil Lawrence , Sheffield

This year, I will serve as an area chair at NIPS 2015 for the first time. Pretty excited and curious to see excellent submissions as well as reviews 🙂

After visiting Google DeepMind, I continued on to visit the University of Oxford. I gave a talk at the <Computational Linguistics Seminars> series about neural machine translation and met many awesome people there. I had interesting discussions with Prof. Nando de Freitas, Prof. Phil Blunsom and their student (Misha Denil, Ziyu Yang and many others).  The next stop was the University of Cambridge thank to Felix Hill and his supervisor, Anna Korhonen. I again presented our neural translation model and my (Felix’s) view/take on syntax to both machine learning researchers and NLP researchers. There were some mixed response, and the

I visited Google DeepMind on Friday (17 Oct). I gave a talk on neural machine translation and met many researchers, including some of my former colleagues such as Razvan Pascanu and Guillaume Desjardins.  As can be expected, I signed NDA before my visit and cannot discuss much about any interesting things that I have heard and discussed at DeepMind. But, I can tell one thing: despite my constant attempt, none of the researchers there was able to tell me a single downside (even when they were quite drunk!). Incredible! Anyways, it was a very interesting and exciting day.