Profile

Cover photo
Yann LeCun
Works at Facebook, Inc.
Attended Université Pierre et Marie Curie
9,668 followers|4,051,406 views
AboutPostsPhotosYouTube

Stream

Yann LeCun
owner

Discussion  - 
 
 
A nice and largely accurate article in The Chronicle of Higher Education about the history of neural nets and deep learning, with quotes from +Geoffrey Hinton, +Terrence Sejnowski, +Yoshua Bengio, and yours truly.

http://chronicle.com/article/The-Believers/190147/
The hidden story behind the code that runs our lives.
View original post
37
18
Massimo Morelli's profile photoEddie Ng's profile photo
Add a comment...

Yann LeCun

Shared publicly  - 
68
12
Madhav Kandukuri's profile photoShawn Simister's profile photoIoannis Kourouklides's profile photoLIU SHR YAN (modeerf)'s profile photo
 
VERY cool. 
Add a comment...

Yann LeCun

Shared publicly  - 
 
Facebook AI Research is open sourcing fbcunn, FAIR's deep learning package for the Torch7 development environment.

This package provides a number of classes and tools for training convolutional nets and other deep learning models. Our library uses super-fast FFT-based convolutions running on NVIDIA GPUs. The package allows training on multiple GPUs. A technical paper gives all the details.

A complete script to train a convolutional net on the ImageNet dataset is provided.

The FAIR blog post announcing the release is here: https://research.facebook.com/blog/879898285375829/fair-open-sources-deep-learning-modules-for-torch/

- Torch website: http://torch.ch/
- fbcunn Github repo: https://github.com/facebook/fbcunn
- ImageNet Training Script: https://github.com/facebook/fbcunn/tree/master/examples/imagenet
- Technical paper on the FFT method used in fbcunn: http://arxiv.org/abs/1412.7580

Our announcement was picked up by quite a few news sites (NY Times, Wired, Gigaom, The Verge, Techcrunch, Venturebeat, ZDNet).

Oren Etzioni (director of the Paul Allen Institute for AI) is quoted in The Verge article: "Whenever you're dealing with a for-profit lab, whether it's Google or Facebook, the question is to what extent will they be part of the academic community and, you know, play nice with others,"

Exactly right, Oren. We very much see ourselves as part of the research community. Research accelerates when people share ideas and tools. We hope that making our tools available will enable brilliant, creative, and fearless young researchers to invent brand new things and push the field forward.

Press coverage:
- NY Times: http://bits.blogs.nytimes.com/2015/01/16/facebook-offers-artificial-intelligence-tech-to-open-source-group
- Gigaom: https://gigaom.com/2015/01/16/facebook-open-sources-tools-for-bigger-faster-deep-learning-models/
- Techcrunch: http://techcrunch.com/2015/01/16/facebook-open-sources-some-of-its-deep-learning-tools/
- Wired: http://www.wired.com/2015/01/facebook-open-sources-trove-ai-tools/
- Venturebeat: http://venturebeat.com/2015/01/16/facebook-opens-up-about-more-of-its-cutting-edge-deep-learning-tools/
- ZDNet: http://www.zdnet.com/article/facebook-open-sources-ai-tools-possibly-turbo-charges-deep-learning/
- The Verge: http://www.theverge.com/2015/1/16/7556691/facebook-artificial-intelligence-research-torch-optimization
The modules are significantly faster than the default ones in Torch and have accelerated research projects by allowing users to train larger neural nets in less time.
108
51
Dennis Sheu's profile photoYili Zhao's profile photoNarek Hovsepyan's profile photoLIU SHR YAN (modeerf)'s profile photo
5 comments
 
Yes, Thank you and the rest of the facebook team for making this happen!
Add a comment...
 
Applications are open for research internships at Facebook AI Research for PhD students. 

FAIR job post: https://www.facebook.com/careers/department?req=a0IA000000CzCGu
FAIR Facebook Page: https://www.facebook.com/FBAIResearch
FAIR Website: https://research.facebook.com/ai
47
12
Mark Bridge's profile photoLukas Zilka's profile photoSayaji Hande's profile photoTheofanis Karaletsos's profile photo
 
Muted....
Add a comment...

Yann LeCun

Shared publicly  - 
 
We are delighted to announce that Vladimir Vapnik has joined Facebook AI Research. 

Vladimir is universally known in the machine learning and statistics communities as the father of statistical learning theory and the co-inventor of the Support Vector Machine method. One of the key concepts of learning theory bears his name: the Vapnik-Chervonenkis Dimension, which measures the capacity of a learning machine.

Vladimir is rejoining some of his long-time collaborators Jason Weston, Ronan Collobert, and Yann LeCun.

He is working on new book, and will be collaborating with FAIR research scientists to develop some of his new ideas on conditional density estimation, learning with privileged information, and other topics.
170
27
Yasser Souri's profile photoCaner Hazırbaş's profile photoDanila Doroshin's profile photoigor tubis's profile photo
9 comments
 
Congrats!!
Add a comment...

Yann LeCun

Shared publicly  - 
 
The October/November issue of the French science magazine "Les Dossiers de la Recherche" has an interview with me about Facebook AI Research, deep learning, and the recent advances in image, speech, and text understanding (page 20). Deputy editor-in-chief Sophie Coisne also devoted her editorial to our encounter last March. 

Much has changed since March. Facebook AI Research now has over 30 scientists and engineers (not 20), and training a convolutional net on ImageNet now takes 2 days, not 2 weeks. 

Pour mes amis français, c'est le numéro 12 de Dossiers avec "La Télévision du Futur" en couverture.

http://www.larecherche.fr/dossiers/12
 ·  Translate
10
4
Emad Barsoum's profile photoIgor Carron's profile photoLukas Mach's profile photoRenaud Chavany's profile photo
2 comments
 
Could I ask how it is possible to learn CNN ImageNet for 2 days? It is multi-GPU, FFT or other stuff (like new GPU)?
Add a comment...
Have him in circles
9,668 people
John Blitzer's profile photo
Aaron Hertzmann's profile photo
Hyunjune Seung's profile photo
Giacomo Indiveri's profile photo
Zvi Kedem's profile photo
Cheong Yuen Kiat's profile photo
gabrutyagi008's profile photo
Funny Clip's profile photo
Shuochen Su's profile photo

Yann LeCun

Shared publicly  - 
 
A nice and largely accurate article in The Chronicle of Higher Education about the history of neural nets and deep learning, with quotes from +Geoffrey Hinton, +Terrence Sejnowski, +Yoshua Bengio, and yours truly.

http://chronicle.com/article/The-Believers/190147/
The hidden story behind the code that runs our lives.
75
57
Eddie Ng's profile photoRaul Mendoza's profile photo
Add a comment...

Yann LeCun
owner

Discussion  - 
 
 
Facebook AI Research is open sourcing fbcunn, FAIR's deep learning package for the Torch7 development environment.

This package provides a number of classes and tools for training convolutional nets and other deep learning models. Our library uses super-fast FFT-based convolutions running on NVIDIA GPUs. The package allows training on multiple GPUs. A technical paper gives all the details.

A complete script to train a convolutional net on the ImageNet dataset is provided.

The FAIR blog post announcing the release is here: https://research.facebook.com/blog/879898285375829/fair-open-sources-deep-learning-modules-for-torch/

- Torch website: http://torch.ch/
- fbcunn Github repo: https://github.com/facebook/fbcunn
- ImageNet Training Script: https://github.com/facebook/fbcunn/tree/master/examples/imagenet
- Technical paper on the FFT method used in fbcunn: http://arxiv.org/abs/1412.7580

Our announcement was picked up by quite a few news sites (NY Times, Wired, Gigaom, The Verge, Techcrunch, Venturebeat, ZDNet).

Oren Etzioni (director of the Paul Allen Institute for AI) is quoted in The Verge article: "Whenever you're dealing with a for-profit lab, whether it's Google or Facebook, the question is to what extent will they be part of the academic community and, you know, play nice with others,"

Exactly right, Oren. We very much see ourselves as part of the research community. Research accelerates when people share ideas and tools. We hope that making our tools available will enable brilliant, creative, and fearless young researchers to invent brand new things and push the field forward.

Press coverage:
- NY Times: http://bits.blogs.nytimes.com/2015/01/16/facebook-offers-artificial-intelligence-tech-to-open-source-group
- Gigaom: https://gigaom.com/2015/01/16/facebook-open-sources-tools-for-bigger-faster-deep-learning-models/
- Techcrunch: http://techcrunch.com/2015/01/16/facebook-open-sources-some-of-its-deep-learning-tools/
- Wired: http://www.wired.com/2015/01/facebook-open-sources-trove-ai-tools/
- Venturebeat: http://venturebeat.com/2015/01/16/facebook-opens-up-about-more-of-its-cutting-edge-deep-learning-tools/
- ZDNet: http://www.zdnet.com/article/facebook-open-sources-ai-tools-possibly-turbo-charges-deep-learning/
- The Verge: http://www.theverge.com/2015/1/16/7556691/facebook-artificial-intelligence-research-torch-optimization
The modules are significantly faster than the default ones in Torch and have accelerated research projects by allowing users to train larger neural nets in less time.
5 comments on original post
43
13
Manu Jeevan Prakash's profile photoTim Moore's profile photoDennis Sheu's profile photoNarek Hovsepyan's profile photo
 
it's great
Add a comment...

Yann LeCun

Shared publicly  - 
 
A crop of new papers submitted to ICLR 2015 by various combinations of my co-authors from Facebook AI Research and NYU.

http://arxiv.org/abs/1412.7580 : "Fast Convolutional Nets With fbfft: A GPU Performance Evaluation" by Nicolas Vasilache, Jeff Johnson, Michael Mathieu, Soumith Chintala, Serkan Piantino, Yann LeCun: two FFT-based implementations of convolutional layers on GPU. The first one is built around NVIDIA's cuFFT, and the second one around custom FFT code called fbfft. This follows our ICLR 2014 paper "Fast Training of Convolutional Networks through FFTs" by Michael Mathieu, Mikael Henaff, Yann LeCun: http://arxiv.org/abs/1312.5851

http://arxiv.org/abs/1412.6651 "Deep learning with Elastic Averaging SGD" by Sixin Zhang, Anna Choromanska, Yann LeCun: a way to distribute the training of deep nets over multiple GPUs by linking the parameter vectors of the workers with "elastics".

http://arxiv.org/abs/1412.7022 "Audio Source Separation with Discriminative Scattering Networks" Pablo Sprechmann, Joan Bruna, Yann LeCun: audio source separation using convnets operating on coefficients of a scattering transform of the signal.

http://arxiv.org/abs/1412.6056 "Unsupervised Learning of Spatiotemporally Coherent Metrics" Ross Goroshin, Joan Bruna, Jonathan Tompson, David Eigen, Yann LeCun: training sparse convolutional auto-encoders so that the pooled features change minimally between successive video frames. Beautiful filters come out of it.

http://arxiv.org/abs/1412.6615 "Explorations on high dimensional landscapes" Levent Sagun, V. Ugur Guney, Yann LeCun: another take on the application of random matrix theory to the geometry of the loss surface in deep nets. We use a "teacher network" and a "student network" scenario to see if SGD manages to find a zero-energy global minimum that we know exists. Bottom line: SGD can't find it, but it doesn't matter because the (local) minima that it finds are equally good as far as test error is concerned.
Abstract: We study the problem of stochastic optimization for deep learning in the parallel computing environment under communication constraints. A new algorithm is proposed in this setting where the communication and coordination of work among concurrent processes (local workers), ...
78
29
Hailin Jin's profile photoDmitry Dmitry's profile photoBrendan Shillingford's profile photoMohammad Havaei's profile photo
2 comments
lus lus
 
great
Add a comment...

Yann LeCun

Shared publicly  - 
 
Facebook AI Research has openings for PhD-level summer interns. 

If you are a PhD student interested in AI, ML, computer vision, NLP, optimization, and related fields, apply here.

https://www.facebook.com/careers/department?req=a0IA000000CzCGu
30
10
Grigory Yaroslavtsev's profile photoArun Iyer's profile photoMichael Witbrock's profile photoBoyu Zhang's profile photo
2 comments
 
Do they experiment on people emotionally?
Add a comment...

Yann LeCun

Shared publicly  - 
 
The NYU Center for Data Science is looking for a new director.

For the last several years, I worked with many colleagues at NYU to create the Center for Data Science. I was the Director of CDS until I stepped down in early 2014 due to my new position at Facebook.

CDS is a major university-wide initiative to promote research and education in the theory and practice of extracting knowledge from data. CDS has a very successful Master's program in Data Science, and is about to start a PhD program in Data Science. The university has allocated a significant amount of resources into the effort, including a (large) number of tenure-track faculty positions and a new building on 5th Avenue. Furthermore, CDS is the recipient of a large 5-year grant from the Moore and Sloan Foundations. 

The NYU CDS is an incredibly exciting initiative that has inspired several institutions around the world to follow suit. 

http://cds.nyu.edu/center-data-science-director/
New York University is seeking to fill the position of Director of its recently created Center for Data Science (CDS). The University is strongly committed to building capacity in Data Science; it has committed to funding new faculty appointments in Data Science and related areas.  CDS has already established a thriving Master’s Program in Data …
14
3
David Kurniawan's profile photoGiacomo Indiveri's profile photoKevin Russell's profile photoCheng Soon Ong's profile photo
 
the link seems to be broken
Add a comment...

Yann LeCun

Shared publicly  - 
 
NVIDIA just announced the new Tesla K80.

The K80 has two GPUs, almost 5000 CUDA cores, 12GB of RAM for each of the two GPUs, and a peak performance is 8.74 teraflops.

With the large RAM we can train larger neural nets, and with the 2 GPUs they will train almost twice as fast.

http://nvidianews.nvidia.com/News/NVIDIA-Unveils-World-s-Fastest-Accelerator-for-Data-Analytics-and-Scientific-Computing-c15.aspx
NVIDIA Tesla K80 Dual-GPU Accelerator Delivers Unmatched Computing Capability With 2x Higher Performance and Memory Bandwidth NVIDIA today unveiled a new addition to the NVIDIA® Tesla Accelerated Computing Platform: the Tesla® K80 dual-GPU accelerator, the world's highest performance accelerator designed for a wide range of machine learning, data analytics, scientific, and high performance computing (HPC) applications.
65
14
Emad Barsoum's profile photoHoang Nguyen's profile photoHassan Hafez's profile photoIoannis Katramados's profile photo
4 comments
 
And in 2016, expected,  they will release their new architecture, Pascal, with 3D memory (so huge memory capacity expected) and NV-link (order of magnitude faster than PCIe) which is a huge step too.  I can see Nvidia is going toward CPUless system!
Add a comment...
People
Have him in circles
9,668 people
John Blitzer's profile photo
Aaron Hertzmann's profile photo
Hyunjune Seung's profile photo
Giacomo Indiveri's profile photo
Zvi Kedem's profile photo
Cheong Yuen Kiat's profile photo
gabrutyagi008's profile photo
Funny Clip's profile photo
Shuochen Su's profile photo
Work
Occupation
Director of AI Research at Facebook & Professor at NYU
Skills
Machine Learning, AI, Computer Perception, Robotics, Research Management
Employment
  • Facebook, Inc.
    Director of AI Research, 2013 - present
  • New York University
    Silver Professor, 2003 - present
  • NEC Labs
    Fellow, 2002 - 2003
  • AT&T Labs
    Research Department Head, 1996 - 2002
  • Bell Labs
    Research Scientist, 1988 - 1996
  • University of Toronto Schools
    Post-Doctoral Research Associate, 1987 - 1988
Basic Information
Gender
Male
Story
Introduction
I work on machine learning, computer vision, robotics, AI, computational neuroscience, and related topics.

I like techno-toys, jazz music, and French wines.

I grew up in Soisy, near Paris, and I have 3 sons.

Bragging rights
I've been known to work on machine learning (deep learning, structured prediction), neural nets (back-prop and convolutional networks), mobile robots (LAGR), image compression (DjVu), handwriting recognition (LeNet), open source software (Lush, EBLearn, djvulibre), and a few other things. I'm also co-founder of MuseAmi Inc, a music technology company.
Education
  • Université Pierre et Marie Curie
    Computer Science, 1984 - 1987
  • ESIEE
    Electrical Engineering, 1978 - 1983