Date: April 11th 2019, 11am
Speaker: Antoine Labatie
Title: Characterizing Well-behaved vs. Pathological Deep Neural Networks [paper]
Room: Kepler
Abstract
We introduce a novel approach, requiring only mild assumptions, for the characterization of deep neural networks at initialization. Our approach applies both to fully-connected and convolutional networks and easily incorporates the commonly used techniques of batch normalization and skip-connections. Our key insight is to consider the evolution with depth of statistical moments of signal and noise, thereby characterizing the presence or the absence of pathologies in the hypothesis space encoded by the choice of hyperparameters. We establish: (i) for feedforward networks with and without batch normalization, depth multiplicativity inevitably leads to ill-behaved moments and pathologies; (ii) for residual networks with batch normalization, on the other hand, skip-connections induce power-law rather than exponential behaviour, leading to well-behaved moments and no pathology.
