Info
Author:
Xavier Glorot
Yoshua Bengio
Year:
2010
Agenda
Abstract
- Deep Neural Networks
- Experimental Setting and Datasets
- Online Learning on an Infinite Dataset: Shapeset-3x2
- Finite Datasets
- Experimental Setting
- Effect of Activation Functions and Saturation During Training
- Experiments with the Sigmoid
- Experiments with the Hyperbolic Tangent
- Experiments with the Softsign
- Studying Gradients and their Propagations
- Effect of the Cost Function
- Gradients at Initialisation
- Theoretical Considerations and a New Normalised Initialisation
- Gradient Propagation Study
- Back-propagation Gradients During
- Error Curves and Conclusions
Evernote!!
https://www.evernote.com/l/ATRQQEYUnXtBTreAuqXoqRWKJjkZn7mNBV0
Nice Material