Open AccessTessarine and Quaternion-Valued Deep Neural Networks for Image Classification
Author(s) -
Fernando Ribeiro de Senna,
Marcos Eduardo Valle
Publication year - 2021
Language(s) - English
Resource type - Conference proceedings
DOI - 10.5753/eniac.2021.18266
Subject(s) - hypercomplex number , quaternion , normalization (sociology) , initialization , computer science , artificial neural network , artificial intelligence , deep learning , pattern recognition (psychology) , deep neural networks , algorithm , mathematics , geometry , sociology , anthropology , programming language
Many image processing and analysis tasks are performed with deep neural networks. Although the vast majority of advances have been made with real numbers, recent works have shown that complex and hypercomplex-valued networks may achieve better results. In this paper, we address quaternion-valued and introduce tessarine-valued deep neural networks, including tessarine-valued 2D convolutions. We also address initialization schemes and hypercomplex batch normalization. Finally, a tessarine-valued ResNet model with hypercomplex batch normalization outperformed the corresponding real and quaternion-valued networks on the CIFAR dataset.