One pixel attack for deceiving deep neural networks

Re:One pixel attack for deceiving deep neural networks by tsomctl on Sunday October 29, 2017 @01:21AM (#15578067)

There are different kinds of attacks. This one, and related ones that minutely change the input, require you to know some details about the model. There are other kinds of attacks, where you don't need to know what the specific model is. For example, http://rocknrollnerd.github.io/ml/2015/05/27/leopard-sofa.ht...

The fundamental problem is that there are only so many training examples and layers that you can fit in a gpu's memory. The current state of the art neural networks work well for very specific tasks (ie handwriting recognition), but don't generalize.

[ Parent | on HN ]

Re:One pixel attack for deceiving deep neural networks by nl on Sunday October 29, 2017 @03:35AM (#15578362)

Neural networks do generalise well, and their vulnerability to crafted attacks is unrelated to generalisation anyway.

Additionally the size of the network isn’t the same thing as GPU memory and you don’t fit training examples in memory anyway (?!).

[ Parent | on HN ]