Don't FREAK Out: A Frequency-Inspired Approach to Detecting Backdoor Poisoned Samples in DNNs

TL;DR

This paper introduces FREAK, a frequency-based detection method for identifying poisoned samples in DNNs, leveraging differences in frequency sensitivity between clean and malicious data.

Contribution

The paper presents a novel frequency analysis approach and a simple yet effective detection algorithm for backdoor poisoned samples in neural networks.

Findings

FREAK effectively detects frequency backdoor attacks.

FREAK also detects some spatial attack poisoned samples.

Significant frequency sensitivity disparities exist between clean and poisoned samples.

Abstract

In this paper we investigate the frequency sensitivity of Deep Neural Networks (DNNs) when presented with clean samples versus poisoned samples. Our analysis shows significant disparities in frequency sensitivity between these two types of samples. Building on these findings, we propose FREAK, a frequency-based poisoned sample detection algorithm that is simple yet effective. Our experimental results demonstrate the efficacy of FREAK not only against frequency backdoor attacks but also against some spatial attacks. Our work is just the first step in leveraging these insights. We believe that our analysis and proposed defense mechanism will provide a foundation for future research and development of backdoor defenses.