Steven Forsyth

William Fleshman, Edward Raff, Jared Sylvester et al.

Adversarial attacks against neural networks are a problem of considerable importance, for which effective defenses are not yet readily available. We make progress toward this problem by showing that non-negative weight constraints can be used to improve resistance in specific scenarios. In particular, we show that they can provide an effective defense for binary classification problems with asymmetric cost, such as malware or spam detection. We also show the potential for non-negativity to be helpful to non-binary problems by applying it to image classification.