发帖
雷达卡
Towards Certifying `∞ Robustness
using Neural Networks with `∞ -dist Neurons
Bohang Zhang 1 Tianle Cai 2 3 Zhou Lu 4 Di He 5 Liwei Wang 1 6
Abstract 1. Introduction
Modern neural networks are usually sensitive to small, ad-
It is well-known that standard neural networks, versarially chosen perturbations to the inputs (Szegedy et al.,
even with a high classification accuracy, are vul- ...
京公网安备 11010802022788号
