arxivst stuff from arxiv that you should probably bookmark

Exploring epoch-dependent stochastic residual networks

Abstract · Apr 20, 2017 15:08 ·

stochastic epoch blocks residual probability cs-cv

Arxiv Abstract

  • Fabio Carrara
  • Andrea Esuli
  • Fabrizio Falchi
  • Alejandro Moreo Fernández

The recently proposed stochastic residual networks selectively activate or bypass the layers during training, based on independent stochastic choices, each of which following a probability distribution that is fixed in advance. In this paper we present a first exploration on the use of an epoch-dependent distribution, starting with a higher probability of bypassing deeper layers and then activating them more frequently as training progresses. Preliminary results are mixed, yet they show some potential of adding an epoch-dependent management of distributions, worth of further investigation.

Read the paper (pdf) »