Hidden Trigger Backdoor Attacks

TL;DR

This paper introduces a novel backdoor attack method that embeds secret triggers in natural-looking poisoned data, effectively fooling models while evading existing defenses.

Contribution

The authors propose a new backdoor attack that hides triggers in natural data and remains undetectable by current defense mechanisms.

Findings

The attack successfully fools models with triggers at random locations.

Models perform well on clean data despite the attack.

Existing defenses cannot reliably detect this new attack.

Abstract

With the success of deep learning algorithms in various domains, studying adversarial attacks to secure deep models in real world applications has become an important research topic. Backdoor attacks are a form of adversarial attacks on deep networks where the attacker provides poisoned data to the victim to train the model with, and then activates the attack by showing a specific small trigger pattern at the test time. Most state-of-the-art backdoor attacks either provide mislabeled poisoning data that is possible to identify by visual inspection, reveal the trigger in the poisoned data, or use noise to hide the trigger. We propose a novel form of backdoor attack where poisoned data look natural with correct labels and also more importantly, the attacker hides the trigger in the poisoned data and keeps the trigger secret until the test time. We perform an extensive study on various image classification settings and show that our attack can fool the model by pasting the trigger at random locations on unseen images although the model performs well on clean data. We also show that our proposed attack cannot be easily defended using a state-of-the-art defense algorithm for backdoor attacks.