Lipschitz Regularization of Neural Networks - Intriguing Properties of Neural Networks

In this video we discuss Lipschitz continuity as a metric for neural network stability to continue with our discussion of robustness to adversarial samples.
Paper: arxiv.org/abs/1312.6199
Abstract:
Deep neural networks are highly expressive models that have recently achieved state of the art performance on speech and visual recognition tasks. While their expressiveness is the reason they succeed, it also causes them to learn uninterpretable solutions that could have counter-intuitive properties. In this paper we report two such properties.First, we find that there is no distinction between individual high level units and random linear combinations of high level units, according to various methods of unit analysis. It suggests that it is the space, rather than the individual units, that contains of the semantic information in the high layers of neural networks.Second, we find that deep neural networks learn input-output mappings that are fairly discontinuous to a significant extend. Specifically, we find that we can cause the network to misclassify an image by applying a certain imperceptible perturbation, which is found by maximizing the network’s prediction error. In addition,the specific nature of these perturbations is not a random artifact of learning: the same perturbation can cause a different network, that was trained on a different subset of the dataset, to misclassify the same input

[@saraghazanfari3480]

Thanks for the material.
You didn't explain the normalization part which is the most tricky.

[@thejudge3094]

Great job! Great explanation

[@MK-sk9wr]

Hi Federico, whats the pdf editor u use?

[@FedericoBarbero00]

Hi Michael, I’m using xournalpp. I’m quite new to this piece of software, my main pdf viewer is Zathura because I like the vim bindings and the fact that it’s very lightweight, but this seems to be quite good for editing PDFs! :-)

[@reutkishon9068]

Hi, I still don't understand how it shows the neural network unstable.. I would be very happy if you could answer 🙏🙏