Fetching the paper…

Spectral regularization for adversarially-robust representation learning · Around