Profile

Cover photo
KyungHyun Cho
497 followers|195,416 views
AboutPostsPhotos+1's

Stream

KyungHyun Cho

Shared publicly  - 
 
The last post in the series on neural machine translation is now available, this time, with some pointers to the actual working code: http://devblogs.nvidia.com/parallelforall/introduction-neural-machine-translation-gpus-part-3/ The timing is good since the ACL has just started. If anyone's attending ACL and wants to discuss about this, feel free to grab me at the conference venue.
17
3
June Hwang's profile photoXu Jia's profile photoMilton Wong's profile photoArthur Chan's profile photo
 
Hi KyungHyun, I just saw that your are pointed as a assistant prof. at NYU. Super congratulations !! I also have been interested in deep learning recently so I just looked your posts from time to time.Just 눈팅. Anyway, lets keep in touch, see you!
Add a comment...

KyungHyun Cho

Shared publicly  - 
 
The panel discussion of the century will be hosted at the Deep Learning Workshop @ICML 2015 in now less than two weeks!
https://sites.google.com/site/deeplearning2015/panel-discussion
18
2
Sungjin Ahn's profile photoKyungHyun Cho's profile photoDurk Kingma's profile photoEmad Barsoum's profile photo
2 comments
 
+Sungjin Ahn it has not be decided yet, and if it'll be, we'll announce it at the workshop website.
Add a comment...

KyungHyun Cho

Shared publicly  - 
 
 
The University of Amsterdam and Qualcomm Technologies are proud to announce the start of a new research lab, QUVA, in deep learning and computer vision.

The agreement has been made for a period of five years and will involve between 15 to 20 researchers. The research lab’s management team will be comprised of Arnold Smeulders, Max Welling, and Cees Snoek.
Please note the availability of 8 PhD positions and 3 postdoc positions. Please apply!

http://www.uva.nl/en/about-the-uva/working-at-the-uva/vacancies/content/2015/06/15-233.html

http://www.uva.nl/en/news-events/news/uva-news/content/press-releases/2015/06/uva-and-qualcom-technologies-inc-to-start-research-lab.html
2 comments on original post
3
2
Grzegorz Chrupała's profile photoSoumith Chintala's profile photo
Add a comment...

KyungHyun Cho

Shared publicly  - 
Abstract: We introduce two Python frameworks to train neural networks on large datasets: Blocks and Fuel. Blocks is based on Theano, a linear algebra compiler with CUDA-support. It facilitates the training of complex neural network models by providing parametrized Theano operations, ...
10
2
Milton Wong's profile photoAlbert Zeyer's profile photo
Add a comment...

KyungHyun Cho

Shared publicly  - 
 
Super interesting paper!
 
We adapted the attention mechanism of the sequence-to-sequence framework to make it possible for the model to output pointers to inputs -- we call this architecture Pointer-Net.  By doing so, we were able to approximately solve problems involving geometry such as small instances of the Traveling Salesman Problem or Delaunay triangulations solely from data.

Our net naturally deals with variable sized problem instances, and generalizes well beyond what’s been trained on. For a video showing how the model generates a triangulation, see https://www.youtube.com/watch?v=yS7rHi_lUGU. For more details, see http://arxiv.org/abs/1506.03134.

+Meire Fortunato +Navdeep Jaitly 
1 comment on original post
6
1
Xu Jia's profile photo
Add a comment...

KyungHyun Cho

Shared publicly  - 
 
Date: June 2, 2015. Dear ILSVRC community,. This is a follow up to the announcement on May 19, 2015 with some more details and the status of the test server. During the period of November 28th, 2014 to May 13th, 2015, there were at least 30 accounts used by a team from Baidu to submit to the ...
7 comments on original post
6
3
Narek Hovsepyan's profile photoGrzegorz Chrupała's profile photo
Add a comment...
Have them in circles
497 people
Jones Kia's profile photo
Francis Quintal Lauzon's profile photo
Kayla Kate's profile photo
Jinhyeon Yun's profile photo
Alvaro Ulloa's profile photo
negar rostamzadeh's profile photo
Sergio Soage's profile photo
Denis Gorodetskiy's profile photo
Bob Dao's profile photo

KyungHyun Cho

Shared publicly  - 
 
The panel discussion of the century will be hosted at the Deep Learning Workshop @ICML 2015 in now less than two weeks! The confirmed speakers now include +Yoshua Bengio, +Neil Lawrence, +Juergen Schmidhuber, +Demis Hassabis, +Yann LeCun and +Kevin Murphy(alphabetical order with two categories. can you guess?) The discussion will be moderated by +Max Welling. Please, leave any question you have about the future of deep learning here by next Wednesday. I will collect the questions and deliver them to Max!
14
Kai Arulkumaran's profile photoAnkur Kumar's profile photoJoan Puigcerver's profile photoKyungHyun Cho's profile photo
7 comments
 
+Joan Puigcerver​ since we need the permission from each and every panelist, I can't say whether it'll be recorded at the moment.
Add a comment...

KyungHyun Cho

Shared publicly  - 
 
"Introduction to Neural Machine Translation", the second part is now up! http://devblogs.nvidia.com/parallelforall/introduction-neural-machine-translation-gpus-part-2/
In my previous post, I introduced statistical machine translation and showed how it can and should be viewed from the perspective of machine learning: as supervised learning where the input and out...
22
14
Benjamin Berhault's profile photoHugo Larochelle's profile photo
Add a comment...

KyungHyun Cho

Shared publicly  - 
 
Didn't think this would work!
 
Are you training recurrent neural networks to produce sequences of tokens (like machine translation or image captioning)?
If so, you should read our recent arxiv paper http://arxiv.org/abs/1506.03099 in which we propose a scheduled sampling approach to improve inference in multi-step prediction tasks by reducing the gap between how you train your model and how you use it for inference.
This is joint work with +Oriol Vinyals , +Navdeep Jaitly , and Noam Shazeer. 
Abstract: Recurrent Neural Networks can be trained to produce sequences of tokens given some input, as exemplified by recent results in machine translation and image captioning. The current approach to training them consists in maximizing the likelihood of each token in the sequence given the ...
2 comments on original post
7
3
Rupesh Kumar Srivastava's profile photo노현우's profile photoNarek Hovsepyan's profile photoAlbert Zeyer's profile photo
 
Didn't know how well it'd work, but discussed exactly this with Quoc at NIPS. We both thought it was definitely worth trying :-) The scheduling might have been the missing link. 
Add a comment...

KyungHyun Cho

Shared publicly  - 
 
 
It is really exciting to see that Montreal/Toronto team takes 3rd place on Microsoft COCO caption generation competition, finishing slightly behind the two giants Google and Microsoft. This is based on the human evaluation results: 

Go to:
http://mscoco.org/dataset/#leaderboard-cap

and click on Table-human.  

Congratulations to +Kelvin Xu , Lei Jimmy, Ba, Ryan Kiros , and +KyungHyun Cho  for pulling this together! 

More interestingly, the model actually uses visual attention! Here is the project page along with the code:

http://kelvinxu.github.io/projects/capgen.html 
Download. Begin by downloading the dataset, including the dataset tools, images, and annotations. Learn about the annotation format. See the cocoDemo in either the Matlab or Python code. demo. Evaluate: Format. Develop your algorithm. Run your algorithm on MS COCO and save the results using the ...
View original post
5
Add a comment...

KyungHyun Cho

Shared publicly  - 
 
NVidia blog post without any code (!) on neural machine translation -- Part 1

http://devblogs.nvidia.com/parallelforall/introduction-neural-machine-translation-with-gpus/
Neural machine translation is a recently proposed framework for machine translation based purely on neural networks. This post is the first of a series in which I will explain a simple encoder-deco...
21
11
Celukuthula Khumalo's profile photoBenjamin Berhault's profile photo
Add a comment...
People
Have them in circles
497 people
Jones Kia's profile photo
Francis Quintal Lauzon's profile photo
Kayla Kate's profile photo
Jinhyeon Yun's profile photo
Alvaro Ulloa's profile photo
negar rostamzadeh's profile photo
Sergio Soage's profile photo
Denis Gorodetskiy's profile photo
Bob Dao's profile photo
Links
KyungHyun Cho's +1's are the things they like, agree with, or want to recommend.
BMW 316 E46 2003 - Used vehicle - Nettiauto
www.nettiauto.com

Nyt myynnissä BMW 316 E46 109 000 km 2003 - Espoo. Klikkaa tästä kuvat ja lisätiedot vaihtoautosta.

InoReader: News and RSS Reader
chrome.google.com

RSS reader built with simplicity and speed in mind. Inspired by Google Reader. Made for Google Reader fans.

ICML 2013 Reading List
peekaboo-vision.blogspot.com

The ICML is now already over for two weeks, but I still wanted to write about my reading list, as there have been some quite interesting pap

The Benefits of Double-Blind Review « Machine Learning (Theory)
hunch.net

This post is a (near) transcript of a talk that I gave at the ICML 2013 Workshop on Peer Review and Publishing Models. Although there's a PD

What it took 20 years on Wall Street to learn | Guest Blog, Scientific A...
blogs.scientificamerican.com

I was always attracted to numbers. In middle school my baseball card collection spilled into a second drawer, driven by my love of the stati

Is It Quantum Computing or Not? | Guest Blog, Scientific American Blog N...
rss.sciam.com

This week I had a fascinating discussion on BBC Radio 4 with Dr Geordie Rose, the CTO of DWave, triggered by the news that NASA and ...

End Piracy, Not Liberty – Google
www.google.com

Millions of Americans oppose SOPA and PIPA because these bills would censor the Internet and slow economic growth in the U.S.. Two bills bef

DNA - Puhelimen tarkemmat tiedot
www.dna.fi

Etusivulle. Sivukartta; Yhteystiedot; In English. Hakusana. Yksityisille; Yrityksille; Nettikauppa; DNA Oy. Puhe; Liikkuva Laajakaista; Laaj

Daum - 생활이 바뀐다! Life On Daum
www.daum.net

전체서비스. 가계부; 검색쇼; 검색트렌드; 공연; 게임; 금융; 날씨; 디렉토리검색; 루리웹; 모바일; 모터쇼; 문자. 문학속세상; 문화유산; 미즈넷; 미즈맘; 미즈쿡; 보안넷. 소설; 소셜쇼핑; 쇼핑원; 쇼핑하우; 스타일; 스포츠. 아이찾기; 여행;

[단독] 박주영, 군대 10년 연기…‘은퇴 후 공익근무요원 복무’ - Daum 스포츠
sports.media.daum.net

[일간스포츠 박린]박주영(28·아스널)이 군대를 미뤘다.박주영 측 대리인인 법무법인 DLS의 이성희 변호사는 16일 "박주영이 국외여행기간 연장 허가 되었음으로 국외 입영연기가 허가됐다. 외국에 거주하면 군대를 미룰 수있다는 허가서를 발급 받았다"고

Docs PDF/PowerPoint Viewer (by Google) - Chrome Web Store
chrome.google.com

Automatically previews pdfs, powerpoint presentations, and other documents in Google Docs Viewer.

Chrome Multitask Mode
www.google.com

Chrome Multitask Mode lets you browse the web with multiple mice at the same time, so you can get more done, clickety-split.

Winter tyre change due soon
www.yle.fi

Approaching colder weather means changing over to winter tyres should be done as soon as possible. According to a survey carried out by Y

News: Two Tenure Track or Tenured Positions in Information and Computer ...
ics.aalto.fi

The positions are open to outstanding individuals who hold a doctorate and have excellent potential for a successful scientific career. Cand

Surplus - Chrome Web Store
chrome.google.com

Google Plus Integration for Chrome

Introducing the ZenRobotics R&D Blog
techblog.zenrobotics.com

Short and sweet: we are launching a blog, by the technical staff, for other technical people, to: Share what it is like to work as an engine

이우진의 이글루 : 줄것처럼 하다가 안주는게 더 싫어
stonepc.egloos.com

줄것처럼 하다가 안주는게 더 싫어. 보통 보면... 얼마 되지도 않는데 오라가라... 이래라 저래라... 그러면서 결제일은 또 늦고.... 그나마 조금이나마 되면 다행인데 뭐 줄것처럼 잔뜩 생색내놓고, 결국엔 "미안합니다 다음 기회에...&q

You're Lucky if You're in Tallinn
online.wsj.com

This charming medieval city is one of Europe's most attractive capitals, with its people still reveling over reestablishing their national i

News: 10 Postdoc / Research Associate positions in the new Finnish Cente...
ics.tkk.fi

Department of Information and Computer Science. Aalto Yliopisto logo. Suomeksi. Contact information; Text version. Department of Information