17.00 - 21.00 Pre-Registration

Morning Session – Session Chair: Dhruv Batra

7.00 - 8.45 Registration

8.45 - 9.00 Opening Remarks [slides]

9.00 - 9.40 Invited talk 1: Eero Simoncelli [slides]

9.40 - 10.00 Contributed talk 1: End-to-end Optimized Image Compression [slides]

10.00 - 10.20 Contributed talk 2: Amortised MAP Inference for Image Super-resolution [slides]

10.20 - 10.30 Coffee Break

10.30 - 12.30 Poster Session 1 (Conference Papers, Workshop Papers)

12.30 - 14.30 Lunch provided by ICLR

Afternoon Session – Session Chair: Joan Bruna (sponsored by Baidu)

14.30 - 15.10 Invited talk 2: Benjamin Recht [slides]

15.10 - 15.30 Contributed Talk 3: Understanding deep learning requires rethinking generalization [slides] - BEST PAPER AWARD

15.30 - 15.50 Contributed Talk 4: Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima [slides]

15.50 - 16.10 Contributed Talk 5: Towards Principled Methods for Training Generative Adversarial Networks [slides]

16.10 - 16.30 Coffee Break

16.30 - 18.30 Poster Session 2 (Conference Papers, Workshop Papers)

Morning Session – Session Chair: Tara Sainath (sponsored by Google)

7.30 – 9.00 Registration

9.00 - 9.40 Invited talk 1: Chloé-Agathe Azencott [slides]

9.40 - 10.00 Contributed talk 1: Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data [slides] - BEST PAPER AWARD

10.00 - 10.20 Contributed talk 2: Learning Graphical State Transitions [slides]

10.20 - 10.30 Coffee Break

10.30 - 12.30 Poster Session 1 (Conference Papers, Workshop Papers)

12.30 - 14.00 Lunch provided by ICLR

Afternoon Session – Session Chair: Raia Hasdell (sponsored by Amazon)

14.00 - 16.00 Poster Session 2 (Conference Papers, Workshop Papers)

16.00 - 16.15 Coffee Break

16.15 - 17.00 Invited talk 2: Riccardo Zecchina [slides]

17.00 - 17.20 Contributed Talk 3: Learning to Act by Predicting the Future [slides]

17.20 - 17.40 Contributed Talk 4: Reinforcement Learning with Unsupervised Auxiliary Tasks [slides]

17.40 - 18.00 Contributed Talk 5: Q-Prop: Sample-Efficient Policy Gradient with An Off-Policy Critic [slides]

18.00 - 18.30 Group photo at the RCT Stadium

19.00 - 24.00 Gala dinner offered by ICLR

The ICLR group photo (1 150 attendees) will be taken in the famous RCT Stadium (Rugby Club Toulonnais, one of the best European team).

We'll use the “secret” door from Neptune congress to get directly into the stadium.

You'll be instructed to arrange in a 14 ranks x 80 people configuration, as illustrated here.

At 18:00, we'll move from the talk session to the reception and then we'll get through the secret door to reach the stadium.

At 18:15, we should be in position, so that we can take the photo by 18:30.

The GALA on 25th in the evening (for conference attendees only) is taking place at Neptune.

There is no dress code for the GALA and admissions are from 19:00 to 19:30. Those wishing to eat earlier can sit on floor 3, where dinner will finish at 22:00. Floor 1 will have dinner served slightly later, finishing at 22:30, but continuing until midnight with live Jazz music.

The menu can be accessed here.

We have a few gluten free and many vegetarian meals for those who asked: the Green Ticket will is for vegetarian meals, whereas the Yellow Ticket is the non-vegetarian option.

The lunches are served at the Neptune congress center. It will consist of finger food, with a few tables for those who wish to sit on floor 0, 1, 2 or 3. Gluten Free meals will be served at floor 2.

Morning Session – Session Chair: Slav Petrov

7.30 – 9.00 Registration

9.00 - 9.40 Invited talk 1: Regina Barzilay [slides]

9.40 - 10.00 Contributed talk 1: Learning End-to-End Goal-Oriented Dialog [slides]

10.00 - 10.20 Contributed talk 2: Multi-Agent Cooperation and the Emergence of (Natural) Language [slides]

10.20 - 10.30 Coffee Break

10.30 - 12.30 Poster Session 1 (Conference Papers, Workshop Papers)

12.30 - 14.30 Lunch provided by ICLR

Afternoon Session – Session Chair: Navdeep Jaitly

14.30 - 15.10 Invited talk 2: Alex Graves [slides]

15.10 - 15.30 Contributed Talk 3: Making Neural Programming Architectures Generalize via Recursion [slides] - BEST PAPER AWARD

15.30 - 15.50 Contributed Talk 4: Neural Architecture Search with Reinforcement Learning [slides]

15.50 - 16.10 Contributed Talk 5: Optimization as a Model for Few-Shot Learning [slides]

16.10 - 16.30 Coffee Break

16.30 - 18.30 Poster Session 2 (Conference Papers, Workshop Papers)

NOTE: No breakfast will be served in the morning.