Fourier Transform Approximation as an Auxiliary Task for Image Classification

TL;DR

This paper explores using Fourier Transform approximation as an auxiliary task in image classification, showing it can improve accuracy and adversarial robustness compared to traditional image reconstruction auxiliary tasks.

Contribution

It introduces Fourier Transform approximation as a novel auxiliary task for image classification, demonstrating its benefits over conventional image reconstruction.

Findings

Improves classification accuracy across multiple architectures

Enhances resistance to adversarial attacks in certain cases

Outperforms traditional auxiliary tasks in experiments

Abstract

Image reconstruction is likely the most predominant auxiliary task for image classification, but we would like to think twice about this convention. In this paper, we investigated "approximating the Fourier Transform of the input image" as a potential alternative, in the hope that it may further boost the performances on the primary task or introduce novel constraints not well covered by image reconstruction. We experimented with five popular classification architectures on the CIFAR-10 dataset, and the empirical results indicated that our proposed auxiliary task generally improves the classification accuracy. More notably, the results showed that in certain cases our proposed auxiliary task may enhance the classifiers' resistance to adversarial attacks generated using the fast gradient sign method.