New Deep Learning Techniques - IPAM

New Deep Learning Techniques

February 5 - 9, 2018

Schedule

All times in this Schedule are Pacific Time (PT)

Monday, February 5, 2018

Morning Session

08:00-08:00
Note: The workshop will be held at the Bruin Reception Room at Ackerman Union (Rm. 2414) on the UCLA campus.
08:00-08:50
Check-In and Breakfast
08:50-09:00
Welcome and Opening Remarks
09:00-09:40
Samuel Bowman (New York University)
Toward natural language semantics in learned representations
09:55-10:10
Break
10:10-10:50
Emily Fox (University of Washington)
Interpretable and Sparse Neural Network Time Series Models for Granger Causality Discovery
11:05-11:20
Break
11:20-12:00
Ellie Pavlick (Brown University)
Should we care about linguistics?

Afternoon Session

12:20-14:00
Lunch (on your own)
14:00-14:40
Leonidas Guibas (Stanford University)
Knowledge Transport Over Visual Data
16:30-17:30
Yann LeCun (New York University)
Public Lecture: Deep Learning and the Future of Artificial Intelligence
17:30-19:00
Poster Session & Reception (Hosted by IPAM)

Tuesday, February 6, 2018

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
09:00-09:40
Alán Aspuru-Guzik (University of Toronto)
Generative models for the inverse design of molecules and materials
09:55-10:10
Break
10:10-10:50
Daniel Rueckert (Imperial College)
Deep learning in medical imaging: Techniques for image reconstruction, super-resolution and segmentation
11:05-11:20
Break
11:20-12:00
Kyle Cranmer (University of Wisconsin-Madison)
Deep Learning in the Physical Sciences

Afternoon Session

12:20-14:00
Lunch (on your own)
14:00-14:40
Stéphane Mallat (École Normale Supérieure)
Deep Generative Networks as Inverse Problems
15:00-15:40
Michael Elad (Technion - Israel Institute of Technology)
Sparse Modeling in Image Processing and Deep Learning
16:30-17:30
Yann LeCun (New York University)
Public Lecture: AI Breakthroughs & Obstacles to Progress, Mathematical and Otherwise
17:30-19:00
Poster Session & Reception (Hosted by IPAM)

Wednesday, February 7, 2018

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
09:00-09:40
Xavier Bresson (National University of Singapore)
Convolutional Neural Networks on Graphs
09:55-10:10
Break
10:10-10:50
Federico Monti (Universita della Svizzera Italiana)
Deep Geometric Matrix Completion: a Geometric Deep Learning approach to Recommender Systems
11:05-11:20
Break
11:20-12:00
Joan Bruna (New York University)
On Computational Hardness with Graph Neural Networks

Afternoon Session

12:20-14:00
Lunch (on your own)
14:00-14:40
Jure Leskovec (Stanford University)
Large-scale Graph Representation Learning
14:55-15:10
Break
15:10-15:50
Arthur Szlam (Facebook)
Composable planning with attributes
16:20-17:00
Yann LeCun (New York University)
A Few (More) Approaches to Unsupervised Learning

Thursday, February 8, 2018

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
09:00-09:40
Sanja Fidler (University of Toronto)
Teaching Machines with Humans in the Loop
09:55-10:10
Break
10:10-10:50
Raquel Urtasun (University of Toronto, Uber)
Deep Learning for Self-Driving Cars
11:05-11:20
Break
11:20-12:00
Pratik Chaudhari (University of California, Los Angeles (UCLA))
Unraveling the mysteries of stochastic gradient descent on deep networks

Afternoon Session

12:20-14:00
Lunch (on your own)
14:00-14:40
Stefano Soatto (University of California, Los Angeles (UCLA))
Emergence Theory of Deep Learning
14:55-15:10
Break
15:10-15:50
Tom Goldstein (University of Maryland)
What do neural net loss functions look like?
16:20-17:00
Stanley Osher (University of California, Los Angeles (UCLA))
New Techniques in Optimization and Their Applications to Deep Learning and Related Inverse Problems

Friday, February 9, 2018

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
09:00-09:40
Michael Bronstein (Imperial College London)
Deep functional maps: intrinsic structured prediction for dense shape correspondence
09:50-10:30
Sainbayar Sukhbaatar (New York University)
Deep Architecture for Sets and Its Application to Multi-agent Communication
10:40-11:00
Break
11:00-11:40
Zuowei Shen (National University of Singapore)
Deep Learning: Approximation of functions by composition
11:50-12:30
Wei Zhu (Duke University)
LDMnet: low dimensional manifold regularized neural networks

Afternoon Session

12:40-12:40
Conclusion
Printable Version