Title | ||
---|---|---|
Efficient Hyperparameter Optimization in Convolutional Neural Networks by Learning Curves Prediction. |
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, prediction of learning curves using non-parametric regression models is applied. Considering that “trend” is the most important feature in any learning curve, our prediction 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. |
Year | Venue | Field |
---|---|---|
2017 | CIARP | Hyperparameter optimization,Hyperparameter,Pattern recognition,Convolutional neural network,Trend detection,Computer science,Regression analysis,Singular spectrum analysis,Artificial intelligence,Deep learning,Learning curve |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
8 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Andrés F. Cardona-Escobar | 1 | 0 | 0.34 |
Andrés Felipe Giraldo-Forero | 2 | 8 | 2.16 |
Andrés Eduardo Castro-Ospina | 3 | 5 | 6.97 |
Jorge Alberto Jaramillo-Garzón | 4 | 17 | 3.96 |