Pyramidal RoR for Image Classification

TL;DR

This paper introduces Pyramidal RoR, a novel image classification network that improves performance by gradually increasing feature map channels, optimizing residual blocks, and using drop-path to prevent overfitting, achieving state-of-the-art results.

Contribution

The paper proposes a Pyramidal RoR model with gradually increasing channels, optimized residual blocks, and drop-path regularization, enhancing RoR performance and training stability.

Findings

Achieved lowest error rates on CIFAR-10/100 and SVHN datasets.

Improved network performance across different datasets.

Effectively mitigated gradient disappearance in training.

Abstract

The Residual Networks of Residual Networks (RoR) exhibits excellent performance in the image classification task, but sharply increasing the number of feature map channels makes the characteristic information transmission incoherent, which losses a certain of information related to classification prediction, limiting the classification performance. In this paper, a Pyramidal RoR network model is proposed by analysing the performance characteristics of RoR and combining with the PyramidNet. Firstly, based on RoR, the Pyramidal RoR network model with channels gradually increasing is designed. Secondly, we analysed the effect of different residual block structures on performance, and chosen the residual block structure which best favoured the classification performance. Finally, we add an important principle to further optimize Pyramidal RoR networks, drop-path is used to avoid over-fitting and save training time. In this paper, image classification experiments were performed on CIFAR-10/100 and SVHN datasets, and we achieved the current lowest classification error rates were 2.96%, 16.40% and 1.59%, respectively. Experiments show that the Pyramidal RoR network optimization method can improve the network performance for different data sets and effectively suppress the gradient disappearance problem in DCNN training.