Adversarial Attacks on Model trained on CIFAR-10
Adversarial attacks on neural networks were developed by Goodfellow et al. as s techniques to fool models through perturbations applied to input data. As a result, the accuracy of the trained model can drop to almost zero. At the same time, perturbed inputs, e.g. images, are imperceptibly different to humans or AI solutions for self-driving cars, the military, or medicine. In this post, I set up a trained model on CIFAR-10. Next, I demonstrate how applied perturbations shift classifications gradually to the false negatives, which can be easily visualised through confusion matrices.
As the medium is not a good platform for scientific blogging, the whole post is available if you click on the sample below. Enjoy reading!
My blog is made in Nikola which seems to be the almost perfect solution for scientific blogging.