Backdoor Mitigation in Deep Neural Networks via Strategic Retraining

TL;DR

This paper presents a new strategy for mitigating backdoors in deep neural networks, especially in safety-critical applications like autonomous driving, without prior knowledge of backdoor characteristics.

Contribution

It introduces a novel backdoor removal method applicable to both malicious and accidental backdoors without needing prior backdoor shape or distribution information.

Findings

Effective removal of backdoors demonstrated on multiple examples

Works for both intentional and unintentional backdoors

No prior knowledge of backdoor properties required

Abstract

Deep Neural Networks (DNN) are becoming increasingly more important in assisted and automated driving. Using such entities which are obtained using machine learning is inevitable: tasks such as recognizing traffic signs cannot be developed reasonably using traditional software development methods. DNN however do have the problem that they are mostly black boxes and therefore hard to understand and debug. One particular problem is that they are prone to hidden backdoors. This means that the DNN misclassifies its input, because it considers properties that should not be decisive for the output. Backdoors may either be introduced by malicious attackers or by inappropriate training. In any case, detecting and removing them is important in the automotive area, as they might lead to safety violations with potentially severe consequences. In this paper, we introduce a novel method to remove backdoors. Our method works for both intentional as well as unintentional backdoors. We also do not require prior knowledge about the shape or distribution of backdoors. Experimental evidence shows that our method performs well on several medium-sized examples.