PCANet: An energy perspective

TL;DR

This paper explains the success of PCANet, a deep learning architecture, from an energy perspective, showing how energy relates to error rates and providing a method to evaluate network steps.

Contribution

It introduces an energy-based explanation for PCANet's effectiveness and proposes a testing approach to assess the necessity of each network component.

Findings

Error rate correlates with the logarithm of image energy.

Energy analysis can identify essential network steps.

Energy perspective offers insights into PCANet's performance.

Abstract

The principal component analysis network (PCANet), which is one of the recently proposed deep learning architectures, achieves the state-of-the-art classification accuracy in various databases. However, the explanation of the PCANet is lacked. In this paper, we try to explain why PCANet works well from energy perspective point of view based on a set of experiments. The impact of various parameters on the error rate of PCANet is analyzed in depth. It was found that this error rate is correlated with the logarithm of energy of image. The proposed energy explanation approach can be used as a testing method for checking if every step of the constructed networks is necessary.