Network Architecture Search Method on Hyperparameter Optimization of Convolutional Neural Network: Review
Hudalizaman, Igi Ardiyanto, Sunu Wibirama · 2020
Deep Learning is a widely used method to solve various kinds of problems. One of the deep learning methods that are used for image classification is Convolutional Neural Network. Convolutional Neural Network has a lot of architecture. Those architectures were manually made and had excellent performance. However, due to a large number of cases tested, it results in the architecture has to be adjusted to the case. If this is manually done, then it should be time-consuming. So there should be an automatic approach to search the appropriate architecture. Neural Architecture Search is one method that is used to search for the appropriate architecture. This paper is containing reviews of Neural Architecture Search method for hyperparameter optimization on Convolutional Neural Network which is done by today's researchers.