A Multiple Classifier Approach for Concatenate-Designed Neural Networks

TL;DR

This paper proposes a multiple classifier approach for concatenate-designed neural networks like ResNet and DenseNet, aiming to improve accuracy and convergence speed by collecting features between network sets and using L2 normalization.

Contribution

The paper introduces a novel multiple classifier method with specific design and normalization techniques that enhance performance and convergence in concatenate-designed neural networks.

Findings

Significant accuracy improvements in experimental cases.

Faster convergence compared to original models.

Applicable to all concatenate-designed network models.

Abstract

This article introduces a multiple classifier method to improve the performance of concatenate-designed neural networks, such as ResNet and DenseNet, with the purpose to alleviate the pressure on the final classifier. We give the design of the classifiers, which collects the features produced between the network sets, and present the constituent layers and the activation function for the classifiers, to calculate the classification score of each classifier. We use the L2 normalization method to obtain the classifier score instead of the Softmax normalization. We also determine the conditions that can enhance convergence. As a result, the proposed classifiers are able to improve the accuracy in the experimental cases significantly, and show that the method not only has better performance than the original models, but also produces faster convergence. Moreover, our classifiers are general and can be applied to all classification related concatenate-designed network models.