Role of Layers and Neurons in Deep Learning With the Rectified Linear Unit

Akira Takekawa; Masayuki Kajiura; Hiroya Fukuda

doi:10.7759/cureus.18866

Role of Layers and Neurons in Deep Learning With the Rectified Linear Unit

Cureus. 2021 Oct 18;13(10):e18866. doi: 10.7759/cureus.18866. eCollection 2021 Oct.

Authors

Akira Takekawa¹, Masayuki Kajiura², Hiroya Fukuda²

Affiliations

¹ Graduate School of Frontier Science, Konan University, Kobe, JPN.
² Graduate School of Human Development and Environment, Kobe University, Kobe, JPN.

Abstract

Deep learning is used to classify data into several groups based on nonlinear curved surfaces. In this paper, we focus on the theoretical analysis of deep learning using the rectified linear unit (ReLU) activation function. Because layers approximate a nonlinear curved surface, increasing the number of layers improves the approximation accuracy of the curved surface. While neurons perform a layer-by-layer approximation of the most appropriate hyperplanes, increasing their number cannot improve the results obtained via canonical correlation analysis (CCA). These results illustrate the functions of layers and neurons in deep learning with ReLU.

Keywords: canonical correlation analysis; curved surface discrimination; deep learning; rectified linear uniit (relu); role of layers; role of neurons; three types of irises.