For some background, see <a href="https://arxiv.org/pdf/1412.1897.pdf" rel="nofollow">https://arxiv.org/pdf/1412.1897.pdf</a> which has some cool visuals of pictures that make neural networks to strange things.<p>Looks like this group might have solved adversarial examples (or rather, found that they're not a problem in practice after all) for self driving cars!<p>Yann LeCun seems confident about it: <a href="https://mobile.twitter.com/ylecun/status/885384106418294784" rel="nofollow">https://mobile.twitter.com/ylecun/status/885384106418294784</a>