The Impact of Uniform Inputs on Activation Sparsity and Energy-Latency Attacks in Computer Vision

arXiv:2403.18587v1 Announce Type: new
Abstract: Resource efficiency plays an important role for machine learning nowadays. The energy and decision latency are two critical aspects to ensure a sustainable and practical application. Unfortunately, the energy consumption and decision latency are not robust against adversaries. Researchers have recently demonstrated that attackers can compute and submit so-called sponge examples at inference time to increase the energy consumption and decision latency of neural networks. In computer vision, the proposed strategy crafts inputs with less …

adversaries application arxiv attacks computer computer vision critical cs.cr cs.cv cs.lg decision efficiency energy impact important inputs latency machine machine learning researchers resource role