@article{Co:2019, author = {Co, KT and Muñoz-González, L and Lupu, EC}, title = {Sensitivity of Deep Convolutional Networks to Gabor Noise}, url = {http://arxiv.org/abs/1906.03455v2}, year = {2019} }
TY - JOUR AB - Deep Convolutional Networks (DCNs) have been shown to be sensitive toUniversal Adversarial Perturbations (UAPs): input-agnostic perturbations thatfool a model on large portions of a dataset. These UAPs exhibit interestingvisual patterns, but this phenomena is, as yet, poorly understood. Our workshows that visually similar procedural noise patterns also act as UAPs. Inparticular, we demonstrate that different DCN architectures are sensitive toGabor noise patterns. This behaviour, its causes, and implications deservefurther in-depth study. AU - Co,KT AU - Muñoz-González,L AU - Lupu,EC PY - 2019/// TI - Sensitivity of Deep Convolutional Networks to Gabor Noise UR - http://arxiv.org/abs/1906.03455v2 ER -