Efficient hyperparameter optimization in convolutional neural networks by learning curves pbkp_rediction

Abstract:

In this work, we present an automatic framework for hyperparameter selection in Convolutional Neural Networks. In order to achieve fast evaluation of several hyperparameter combinations, pbkp_rediction of learning curves using non-parametric regression models is applied. Considering that “trend” is the most important feature in any learning curve, our pbkp_rediction method is focused on trend detection. Results show that our forecasting method is able to catch a complete behavior of future iterations in the learning process.

Año de publicación:

2018

Keywords:

• forecasting
• Learning curves
• deep learning
• Singular spectrum analysis
• SVR
• Hyperparameter optimization

Fuente:

scopus