Graduate Summer School: Deep Learning, Feature Learning - IPAM

Graduate Summer School: Deep Learning, Feature Learning

July 9 - 27, 2012

Schedule

All times in this Schedule are Pacific Time (PT)

Monday, July 9, 2012

Morning Session

08:00-08:45
Check-In/Light Breakfast (Hosted by IPAM)
08:45-09:00
Welcome and Opening Remarks
09:00-10:00
Geoffrey Hinton (University of Toronto)
PART 1: Introduction to Deep Learning & Deep Belief Nets (PDF Parts 1 & 2)
10:00-10:30
Break
10:30-11:30
Geoffrey Hinton (University of Toronto)
PART 2: Using backpropagation for fine-tuning a generative model to be better at discrimination
11:30-12:00
Break

Afternoon Session

12:00-13:00
Joint talk: James Bergstra (Harvard), Clement Farabet (NYU)
13:00-14:30
Lunch (on your own)
14:30-15:30
Yann LeCun (New York University)
Deep Learning, Graphical Models, Energy-Based Models, Structured Prediction
15:30-16:00
Break
16:00-17:00
Yann LeCun (New York University)
Deep Learning, Graphical Models, EnergyBased Models, Structured Prediction (Part 2)
17:00-18:30
Reception (Location: IPAM Lobby)

Tuesday, July 10, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Andrew Ng (Stanford University)
Deep Learning, Self-Taught Learning and Unsupervised Feature Learning (Part 1 Slides1-68; Part 2 Slides 69-109)
10:00-10:30
Break
10:30-11:30
Andrew Ng (Stanford University)
Advanced topics + Research philosophy / Neural Networks: Representation
11:30-12:00
Break

Afternoon Session

12:00-13:00
Andrew Ng (Stanford University)
Non-linear hypotheses
13:00-14:30
Lunch (on your own)
14:30-15:30
Andrew Ng (Stanford University)
Non-linear hypotheses (Part 2)
15:30-16:00
Break
16:00-17:00
Joint talk: James Bergstra (Harvard), Clement Farabet (NYU)

Wednesday, July 11, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Yann LeCun (New York University)
Learning Hierarchies of Invariant Features (Parts 1 & 2)
10:00-10:30
Break
10:30-11:30
Yann LeCun (New York University)
Deep Learning, Graphical Models, EnergyBased Models, Structured Prediction (Part 3)
11:30-12:00
Break

Afternoon Session

12:00-13:00
Joint talk: James Bergstra (Harvard), Clement Farabet (NYU)
13:00-14:30
Lunch (on your own)
14:30-15:30
Geoffrey Hinton (University of Toronto)
PART 3: Some applications of deep learning (Slides 1-38)
15:30-16:00
Break
16:00-17:00
Geoffrey Hinton (University of Toronto)
PART 4: A computational principle that explains sex, the brain, and sparse coding (Slides 39-92)

Thursday, July 12, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Rob Fergus (New York University)
Deep Learning Methods for Vision
10:00-10:30
Break
10:30-11:30
Rob Fergus (New York University)
Deep Learning Methods for Vision (Part 2)
11:30-12:00
Break

Afternoon Session

12:00-13:00
Geoffrey Hinton (University of Toronto)
Title not available
13:00-14:30
Lunch (on your own)
14:30-15:30
Alan Yuille (University of California, Los Angeles (UCLA))
Part 1
15:30-16:00
Break
16:00-17:00
Joint talk: James Bergstra (Harvard), Clement Farabet (NYU)

Friday, July 13, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Graham Taylor (University of Guelph)
Title not available
10:00-10:30
Break
10:30-11:30
Graham Taylor (University of Guelph)
Title not available
11:30-12:00
Break

Afternoon Session

12:00-13:00
Alan Yuille (University of California, Los Angeles (UCLA))
Part 2
13:00-14:30
Lunch (on your own)
14:30-15:30
Alan Yuille (University of California, Los Angeles (UCLA))
Part 3
15:30-16:00
Break
16:00-17:00
Yann LeCun (New York University)
Deep Learning, Graphical Models, EnergyBased Models, Structured Prediction (Part 4)

Monday, July 16, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Yoshua Bengio (University of Montreal)
Part 1
10:00-10:30
Break
10:30-11:30
Yoshua Bengio (University of Montreal)
Part 2
11:30-12:00
Break

Afternoon Session

12:00-13:00
Stephen Wright (University of Wisconsin-Madison)
Some Relevant Topics in Optimization (Part 1) - (Slides Cover Parts 1 & 2)
13:00-14:30
Lunch (on your own)
14:30-15:30
Stephen Wright (University of Wisconsin-Madison)
Some Relevant Topics in Optimization (Part 2)
15:30-16:00
Break
16:00-17:00
Arthur Szlam (Facebook)
A tutorial on sparse modeling.
17:00-18:30
Reception (Location: IPAM Lobby)

Tuesday, July 17, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Stephen Wright (University of Wisconsin-Madison)
Sparse and Regularized Optimization Part 1 (Slides Cover Parts 1 & 2)
10:00-10:30
Break
10:30-11:30
Stephen Wright (University of Wisconsin-Madison)
Sparse and Regularized Optimization Part 2
11:30-12:00
Break

Afternoon Session

12:00-13:00
Arthur Szlam (Facebook)
Title not available
13:00-14:30
Lunch (on your own)
14:30-15:30
Yoshua Bengio (University of Montreal)
Part 3
15:30-16:00
Break
16:00-17:00
Yoshua Bengio (University of Montreal)
Part 4

Wednesday, July 18, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Stéphane Mallat (École Polytechnique)
Scattering Invariant Deep Networks for Classification
10:00-10:30
Break
10:30-11:30
Stéphane Mallat (École Polytechnique)
Scattering Invariant Deep Networks for Classification (Part 2)
11:30-12:00
Break

Afternoon Session

12:00-13:00
Kai Yu (Baidu Inc.)
Part 1
13:00-14:30
Lunch (on your own)
14:30-15:30
Kai Yu (Baidu Inc.)
Part 2
15:30-16:00
Break
16:00-17:00
Jorge Nocedal (Northwestern University)
Tutorial on Optimization methods for machine learning

Thursday, July 19, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Jorge Nocedal (Northwestern University)
Tutorial on Optimization methods for machine learning (Part 2)
10:00-10:30
Break
10:30-11:30
Jorge Nocedal (Northwestern University)
Tutorial on Optimization methods for machine learning (Part 3)
11:30-14:00
Lunch (on your own)

Afternoon Session

14:00-15:30
Stéphane Mallat (École Polytechnique)
Scattering Invariant Deep Networks for Classification (Part 3)
15:30-16:00
Break
16:00-17:00
Jason Morton (Pennsylvania State University)
An Algebraic Perspective on Deep Learning

Friday, July 20, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Jason Morton (Pennsylvania State University)
An Algebraic Perspective on Deep Learning (Part 2)
10:00-10:30
Break
10:30-11:30
Jason Morton (Pennsylvania State University)
An Algebraic Perspective on Deep Learning (Part 3)
11:30-12:00
Break

Afternoon Session

12:00-13:00
Yoshua Bengio (University of Montreal)
Title not available
13:00-14:30
Lunch (on your own)
14:30-15:30
Stanley Osher (University of California, Los Angeles (UCLA))
Title not available
15:30-16:00
Break
16:00-17:00
Stanley Osher (University of California, Los Angeles (UCLA))
Title not available

Monday, July 23, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Ruslan Salakhutdinov (University of Toronto)
Title not available
10:00-10:30
Break
10:30-11:30
Ruslan Salakhutdinov (University of Toronto)
Title not available
11:30-12:00
Break

Afternoon Session

12:00-13:00
Marc'Aurelio Ranzato (Google Inc.)
Deep Gated MRF's - part 1
13:00-14:30
Lunch (on your own)
14:30-15:30
Marc'Aurelio Ranzato (Google Inc.)
Deep Gated MRF's - part 2
15:30-16:00
Break
16:00-17:00
Jason Weston (Facebook)
Part 1
17:00-18:30
Reception (Location: IPAM Lobby)
18:30-19:30
Brian Milch (Google Inc.)
From Text to Concepts at Google

Tuesday, July 24, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Jason Weston (Facebook)
Part 2
10:00-10:30
Break
10:30-11:30
Jason Weston (Facebook)
Part 3
11:30-12:00
Break

Afternoon Session

12:00-13:00
Ruslan Salakhutdinov (University of Toronto)
Title not available
13:00-14:30
Lunch (on your own)
14:30-15:30
Bruno Olshausen (University of California, Berkeley (UC Berkeley))
From natural scene statistics to models of neural coding and representation (part 1)
15:30-16:00
Break
16:00-17:00
Marc'Aurelio Ranzato (Google Inc.)
Large Scale Deep Learning

Wednesday, July 25, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Thomas Serre (Brown University)
Deep learning in the visual cortex - Part 1
10:00-10:30
Break
10:30-11:30
Thomas Serre (Brown University)
Deep learning in the visual cortex - Part 2
11:30-12:00
Break

Afternoon Session

12:00-13:00
Roland Memisevic (Johann Wolfgang Goethe-Universität Frankfurt)
Multiview Feature Learning - Part 1
13:00-14:30
Lunch (on your own)
14:30-15:30
Roland Memisevic (Johann Wolfgang Goethe-Universität Frankfurt)
Multiview Feature Learning - Part 2
15:30-16:00
Break
16:00-17:00
Bruno Olshausen (University of California, Berkeley (UC Berkeley))
From natural scene statistics to models of neural coding and representation (part 2)

Thursday, July 26, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Iain Murray (University of Edinburgh)
Introduction to MCMC for deep learning
10:00-10:30
Break
10:30-11:30
Iain Murray (University of Edinburgh)
Density estimation
11:30-12:00
Break

Afternoon Session

12:00-13:00
Nando de Freitas (University of British Columbia)
Part 1
13:00-14:30
Lunch (on your own)
14:30-15:30
Nando de Freitas (University of British Columbia)
Part 2
15:30-16:00
Break
16:00-17:00
Thomas Serre (Brown University)
Deep learning in the visual cortex - Part 3

Friday, July 27, 2012

Morning Session

08:00-09:00
Continental Breakfast
09:00-10:00
Nando de Freitas (University of British Columbia)
An Informal Mathematical Tour of Feature Learning
10:00-10:30
Break
10:30-11:30
Nando de Freitas (University of British Columbia)
Part 3

Afternoon Session

12:00-13:30
Lunch (on your own)
13:30-14:30
Panel Discussion: Nando de Freitas, Iain Murray, Bruno Olshausen, Ruslan Salakhutdinov, Roland Memisevic
14:30-15:00
Closing Remarks by Russ Caflisch (IPAM Director)
Printable Version