Schedule

Invited Talks
  • Yoshua Bengio: Towards disentangling underlying explanatory factors
  • Taco Cohen: The General Theory of Equivariant Convolutional Networks
  • Nicholas Frosst: Capsule networks and transformation extrapolation for learning from limited data
  • Dinesh Jayaraman: Towards Embodied Vision
  • Tomaso Poggio: A surprising linear relationship predicts test performance in Deep Networks
  • Stefano Soatto: The Emergence of Invariance and Independence in Representation Learning


8:30 - 8:50

Poster setup

8:50 - 9:00

Opening remarks

9:00 - 9:30

Taco Cohen (Invited talk)

9:30 - 9:45

Learning representations that account for data symmetries (Contributed talk)

9:45 - 10:00

Scale equivariance in CNNs with vector fields (Contributed talk)

10:00 - 10:30

Coffee break

10:30 - 11:00

Tomaso Poggio (Invited talk)

11:00 - 11:30

Nicholas Frosst (Invited talk)

11:30 - 11:45

Planning with Arithmetic and Geometric Attributes (Contributed talk)

11:45 - 12:00

A Kernel Theory of Modern Data Augmentation (Contributed talk)

12:00 - 14:00

Lunch

14:00 - 14:30

Yoshua Bengio (Invited talk)

14:30 - 14:45

Universal approximations of invariant maps by neural networks (Contributed talk)

14:45 - 15:00

3D Group-Equivariant Neural Networks for Octahedral and Square Prism Symmetry Groups (Contributed talk)

15:00-16:30

Poster Session

16:30 - 17:00

Stefano Soatto (Invited talk)

17:00 - 17:30

Dinesh Jayaraman (Invited talk)

17:30

Closing remarks