This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
Next revision Both sides next revision
adversarial_features [2018/06/20 18:16]
adversarial_features [2018/07/11 10:37]
Line 259: Line 259:
 https://​arxiv.org/​abs/​1806.06108v1 Non-Negative Networks Against Adversarial Attacks https://​arxiv.org/​abs/​1806.06108v1 Non-Negative Networks Against Adversarial Attacks
 +https://​arxiv.org/​pdf/​1805.12177v1.pdf Why do deep convolutional networks generalize so
 +poorly to small image transformations?​
 +https://​arxiv.org/​abs/​1806.11146 Adversarial Reprogramming of Neural Networks
 +https://​arxiv.org/​pdf/​1805.12152.pdf There Is No Free Lunch In Adversarial Robustness
 +(But There Are Unexpected Benefits)
 +An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution
 +https://​arxiv.org/​abs/​1807.03247 ​