Enhancing ResNet Image Classification Performance by using Parameterized Hypercomplex Multiplication

TL;DR

This paper investigates the impact of integrating parameterized hypercomplex multiplication into the backend of ResNet architectures, demonstrating improved classification accuracy across various image datasets and achieving state-of-the-art results for hypercomplex networks.

Contribution

It introduces the novel application of PHM to the backend of ResNet architectures, expanding hypercomplex methods beyond the convolutional frontend.

Findings

PHM improves classification accuracy on CIFAR 10/100, ImageNet, and ASL datasets.

Incorporating PHM achieves state-of-the-art results for hypercomplex networks.

Enhancement is effective for both small and large-scale image classification tasks.

Abstract

Recently, many deep networks have introduced hypercomplex and related calculations into their architectures. In regard to convolutional networks for classification, these enhancements have been applied to the convolution operations in the frontend to enhance accuracy and/or reduce the parameter requirements while maintaining accuracy. Although these enhancements have been applied to the convolutional frontend, it has not been studied whether adding hypercomplex calculations improves performance when applied to the densely connected backend. This paper studies ResNet architectures and incorporates parameterized hypercomplex multiplication (PHM) into the backend of residual, quaternion, and vectormap convolutional neural networks to assess the effect. We show that PHM does improve classification accuracy performance on several image datasets, including small, low-resolution CIFAR 10/100 and large high-resolution ImageNet and ASL, and can achieve state-of-the-art accuracy for hypercomplex networks.