SwGridNet: A Deep Convolutional Neural Network based on Grid Topology for Image Classification

TL;DR

SwGridNet introduces a grid topology-based deep CNN architecture with multiple processing paths, enhancing generalization and achieving competitive image classification performance on CIFAR datasets.

Contribution

The paper proposes SwGridNet, a novel grid-structured CNN that leverages multipath architecture to improve generalization in image classification.

Findings

Achieves 2.95% test error on CIFAR-10

Achieves 15.67% test error on CIFAR-100

Performance comparable to state-of-the-art deep CNNs

Abstract

Deep convolutional neural networks (CNNs) achieve remarkable performance on image classification tasks. Recent studies, however, have demonstrated that generalization abilities are more important than the depth of neural networks for improving performance on image classification tasks. Herein, a new neural network called SwGridNet is proposed. A SwGridNet includes many convolutional processing units which connect mutually as a grid network where many processing paths exist between input and output. A SwGridNet has high generalization capability because the multipath architecture has the same effect of ensemble learning. As described in this paper, details of the SwGridNet network architecture are presented. Experimentally obtained results presented in this paper show that SwGridNets respectively achieve test error rates of 2.95% and 15.67% in a CIFAR-10 and CIFAR-100 classification tasks. The results indicate that the SwGridNet performance approximates that of state-of-the-art deep CNNs.