Code supporting the publication: BAN: detecting backdoors activated by adversarial neuron noise.

This repo contains the code for Xu et al. "BAN: detecting backdoors activated by adversarial neuron noise." Advances in Neural Information Processing Systems 37 (2024): 114348-114373. This paper improves backdoor feature inversion for backdoor detection by incorporating extra neuron activation information. We adversarially increase the loss of backdoored models with respect to weights to activate the backdoor effect, based on which we can easily differentiate backdoored and clean models.