Input-Aware Dynamic Backdoor Attack

Understand

In recent years, neural backdoor attack has been considered to be a potential security threat to deep learning systems.

• Such systems, while achieving the state-of-the-art performance on clean data, perform abnormally on inputs with predefined triggers.
• Current backdoor techniques, however, rely on uniform trigger patterns, which are easily detected and mitigated by current defense methods.
• In this work, we propose a novel backdoor attack technique in which the triggers vary from input to input.