no code implementations • 10 Nov 2021 • Giuseppina Carannante, Dimah Dera, Nidhal C. Bouaynaya, Hassan M. Fathallah-Shaykh, Ghulam Rasool
Moreover, the uncertainty map of the proposed SUPER-Net associates low confidence (or equivalently high uncertainty) to patches in the test input images that are corrupted with noise, artifacts, or adversarial attacks.
no code implementations • 10 Nov 2021 • Giuseppina Carannante, Dimah Dera, Ghulam Rasool, Nidhal C. Bouaynaya, Lyudmila Mihaylova
Learning in uncertain, noisy, or adversarial environments is a challenging task for deep neural networks (DNNs).
no code implementations • 10 Nov 2021 • Giuseppina Carannante, Dimah Dera, Ghulam Rasool, Nidhal C. Bouaynaya
We show that Bayesian neural networks automatically discover redundancy in model parameters, thus enabling self-compression, which is linked to the propagation of uncertainty through the layers of the network.