发帖
雷达卡
Improving Gradient Regularization
using Complex-Valued Neural Networks
Eric Yeats 1 Yiran Chen 1 Hai Li 1
Abstract and that these inputs are difficult to distinguish from nor-
Gradient regularization is a neural network de- mal inputs. Subsequent investigation by Goodfellow et al.
fense technique that requires no prior knowledge (2014) and Papernot et al. (2016) further demonstrate that
of an adversarial attack ...
京公网安备 11010802022788号
