Apply for community grant: Academic project (gpu)

#1
by rgrainger - opened

QuadAttack is an adversarial attack on image classification backbones. It allows the attacker to specify any top K classes they desire the target network to output with no human-visible change to the input image. This demo runs QuadAttack on an image, and outputs the original classifications, the resulting classification as well as the input image, attacked image, and the noise the network generated. This demo is based on the QuadAttack paper accepted into NeurIPS 2023.

Hi @rgrainger , we have assigned a gpu to this space. Note that GPU Grants are provided temporarily and might be removed after some time if the usage is very low.

To learn more about GPUs in Spaces, please check out https://huggingface.co/docs/hub/spaces-gpus

Sign up or log in to comment