Program

The meeeting will have four research-oriented sessions plus two sessions on applications, according to the schedule below. There will be a welcome reception on Sunday evening at 20:00.

Time

Monday Morning: NEUROBIOLOGY AND THEORY
Chair: Tomaso Poggio

Tuesday Morning: ALGORITHMS 1
Chair: Paolo Frasconi

Wednesday Morning: APPLICATIONS 1
Chair: Sepp Hochreiter

Thursday Morning: APPLICATIONS 2
Chair: Marco Gori

9:00

Marco Zorzi: Deep Learning Models of Perception and Coognition

Rich Caruana: Do Deep Nets Really Need To Be Deep?


Soeren Brunak: Heterogeneous Data from the Healthcare Field and their Analysis


9:20

Pierre Baldi: Deep Learning, Dark Knowledge, Dark Matter

9:40

Wolfgang Maass: Why the Cortex Would Have Trouble Running Currently Considered Deep Learning   Algorithms, and what Alternative Tricks it Could Possibly Use

Francesco Palmieri: Factor Graphs in Reduced Normal Form for building distributed Belief Architectures


Pietro Di Lena: Deep Architectures for Protein Contact Prediction

10:00

Gianluca Pollastri: Deep Learning for Predicting Structural Features of Proteins

Sepp Hochreiter: Toxicity Prediction using Deep Learning

10:20

Coffee break

Coffee break

Coffee break

Coffee break

10:50

Thomas Serre: Towards a System-Level Theory of Computation in the Visual Cortex

Eric Mjolsness: Update on Symbolic Quantitative Modeling Languages and Methods

Piero Fariselli: Machine Learning Methods for the Detection and Prediction of Transmembrane Beta-Barrel Proteins in Prokariotes

Maxime Oquab: Is Object Localization for Free? – Weakly-Supervised Learning with Convolutional Neural Networks

11:10

Pierangelo Veltri: Bioinformatics Applied to SmallRNA and RNA-seq: Application to NGS Data

Paolo Frasconi: Algorithms for mapping neuronal activity with cellular resolution on a brain-wide scale

11:30

Tomaso Poggio: Learning of Invariant Representations in Visual Cortex: i-Theory

Vijay Saraswat: Deep Learning in Resilient X10

Pierangelo Veltri: Machine Learning Techniques for Analysis and Monitoring of Clinical and Environmental Issues Related to Life Quality and Public Health

Discussion

11:50

Marco Zorzi: Modeling Human Developmental Trajectories with Deep Learning

Panel

12:10


Guillermo Sapiro: Random Forests can Hash (and Outperform Deep Learning)

12:30

Lunch

Lunch

Lunch

Lunch


Monday Afternoon: NEUROBIOLOGY AND THEORY
Chair: Pierre Baldi

Tuesday Afternoon: ALGORITHMS 2
Chair: Alessandro Sperduti



14:30

Pierre Baldi: The Ebb and Flow of Deep Learning: a Theory of Local Learning

Sepp Hochreiter: Rectified Factor Networks

Excursion + Dinner


14:50

Sepp Hochreiter: Long Short-Term Memory for Uniform Credit Assignment

15:10

Bernardo Sabatini: Neurotransmitter co-Release in the Brain

Stephane Mallat: Geometric Invariants with Deep Scattering Networks: From Image Recognition to Quantum Chemistry

15:30

15:50

Coffee break

Coffee break

16:20

Guillermo Sapiro: On the Stability of Deep Networks

Marcello Pelillo: How Mature Is the Field of Machine Learning?

16:40

Alessandro Sperduti: Equivalence Results between Feedforward and Recurrent Neural Networks for Sequences

17:00

Marco Gori: The Principle of Least Cognitive Action

Lorenzo Rosasco: Convolution Layers Build Invariant and Selective Reproducing Kernels

17:20

Panel

Panel

17:40

18:00

Free time

Free time

20:00

Dinner

Dinner