Representation Learning

Program

Foundations of Machine Learning

Location

Calvin Lab Auditorium

Date

Monday, Mar. 27 – Friday, Mar. 31, 2017

Back to calendar

9 – 9:20 a.m.

Coffee and Check-In

9:20 – 9:30 a.m.

Opening Remarks

9:30 – 10:10 a.m.

Geometry, Optimization and Generalization in Multilayer Networks

Nathan Srebro, TTI Chicago

10:10 – 10:50 a.m.

A Non-generative Framework and Convex Relaxations for Unsupervised Learning

Elad Hazan, Princeton University

10:50 – 11:20 a.m.

Break

11:20 a.m. – 12:20 p.m.

Representations for Language: From Word Embeddings to Sentence Meanings

Christopher Manning, Stanford University

12:20 – 2:15 p.m.

Lunch

2:15 – 3:15 p.m.

The Missing Signal

Leon Bottou, Facebook AI Research

3:15 – 3:45 p.m.

Break

3:45 – 4:25 p.m.

Representations of Relationships in Images and Text

Hal Daume, University of Maryland at College Park

4:30 – 4:45 p.m.

Spotlight Talk: A Formalization of Representation Learning

Andrej Risteski, Princeton University

4:45 – 5 p.m.

Spotlight Talk: Semi-Random Units for Learning Neural Networks with Guarantees

Bo Xie, Georgia Institute of Technology

5 – 6 p.m.

Reception

9 – 9:30 a.m.

Coffee and Check-In

9:30 – 10:10 a.m.

Unsupervised Discovery Through Adversarial Self-Play

Rob Fergus, New York University

10:10 – 10:50 a.m.

Learning from Unlabeled Video

Kristen Grauman, University of Texas, Austin

10:50 – 11:20 a.m.

Break

11:20 a.m. – 12:20 p.m.

Deep Reinforcement Learning

Pieter Abbeel, UC Berkeley

12:20 – 2:15 p.m.

Lunch

2:15 – 2:55 p.m.

Supersizing Self-Supervision: Learning Perception and Action without Human Supervision

Abhinav Gupta, Carnegie Mellon University

2:55 – 3:35 p.m.

Failures of Deep Learning

Shai Shalev-Shwartz, Hebrew University of Jerusalem

3:35 – 4:10 p.m.

Break

4:10 – 5 p.m.

Panel

9 – 9:30 a.m.

Coffee and Check-In

9:30 – 10:10 a.m.

Continuous State Machines and Grammars for Linguistic Structure Prediction

Noah Smith, University of Washington

10:10 – 10:50 a.m.

Adversarial Perceptual Representation Learning Across Diverse Modalities and Domains

Trevor Darrell, UC Berkeley

10:50 – 11:20 a.m.

Break

11:20 a.m. – 12:20 p.m.

Evaluating Neural Network Representations Against Human Cognition

Tom Griffiths, UC Berkeley

12:20 – 2:15 p.m.

Lunch

2:15 – 2:55 p.m.

Representation Learning for Reading Comprehension

Russ Salakhutdinov, Carnegie Mellon University

2:55 – 3:35 p.m.

Tractable Learning in Structured Probability Spaces

Adnan Darwiche, UCLA

3:35 – 4:10 p.m.

Break

4:10 – 4:30 p.m.

Spotlight Talk: Convolutional Dictionary Learning through Tensor Factorization

Furong Huang, UC Irvine

4:35 – 4:55 p.m.

Spotlight Talk: How to Escape Saddle Points Efficiently

Praneeth Netrapalli, Microsoft Research India

9 – 9:30 a.m.

Coffee and Check-In

9:30 – 10:10 a.m.

Re-Thinking Representational Learning in Robotics and Music

Sham Kakade, University of Washington

10:10 – 10:50 a.m.

Representation Learning of Grounded Language and Knowledge: with and without End-to-End Learning

Yejin Choi, University of Washington

10:10 – 10:50 a.m.

Representation Learning of Grounded Language and Knowledge: with and without End-to-End Learning

Yejin Choi, University of Washington

10:50 – 11:20 a.m.

Break

11:20 a.m. – 12:20 p.m.

TBD

Yann LeCun, New York University

12:20 – 2:15 p.m.

Lunch

2:15 – 2:55 p.m.

Generalization and Equilibrium in Generative Adversarial Nets (GANs)

Sanjeev Arora, Princeton University

2:55 – 3:35 p.m.

Provably Learning of Noisy-or Networks

Rong Ge, Duke University

3:35 – 4:10 p.m.

Break

4:10 – 5 p.m.

Panel

9 – 9:30 a.m.

Coffee and Check-In

9:30 – 10:10 a.m.

Learning Paraphrastic Representations of Natural Language Sentences

Kevin Gimpel, TTI Chicago

10:10 – 10:50 a.m.

Formation and Association of Symbolic Memories in the Brain

Christos Papadimitriou, UC Berkeley

10:50 – 11:20 a.m.

Break

11:20 a.m. – 12:20 p.m.

Learning Representations for Active Vision

Bruno Olshausen, UC Berkeley

12:20 – 2:15 p.m.

Lunch

2:15 – 2:55 p.m.

Resilient Representation and Provable Generalization

Dawn Song, UC Berkeley

2:55 – 3:35 p.m.

Word Representation Learning without <unk> Assumptions

Chris Dyer, Carnegie Mellon University

3:35 – 4:10 p.m.

Break

4:10 – 4:30 p.m.

Spotlight Talk: Performance Guarantees for Transferring Representations

Daniel McNamara, Australian National University