SHADE: Information-Based Regularization for Deep Learning - Sorbonne Université Access content directly
Conference Papers Year : 2018

SHADE: Information-Based Regularization for Deep Learning


Regularization is a big issue for training deep neural networks. In this paper, we propose a new information-theory-based regularization scheme named SHADE for SHAnnon DEcay. The originality of the approach is to define a prior based on conditional entropy, which explicitly decouples the learning of invariant representations in the regularizer and the learning of correlations between inputs and labels in the data fitting term. Our second contribution is to derive a stochastic version of the regularizer compatible with deep learning, resulting in a tractable training scheme. We empirically validate the efficiency of our approach to improve classification performances compared to common regularization schemes on several standard architectures.
Fichier principal
Vignette du fichier
1804.10988.pdf (604.78 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-01994740 , version 1 (25-01-2019)



Michael Blot, Thomas Robert, Nicolas Thome, Matthieu Cord. SHADE: Information-Based Regularization for Deep Learning. ICIP 2018 - 25th IEEE International Conference on Image Processing, Oct 2018, Athènes, Greece. pp.813-817, ⟨10.1109/ICIP.2018.8451092⟩. ⟨hal-01994740⟩
102 View
236 Download



Gmail Mastodon Facebook X LinkedIn More