Robust adversarial inputs
OpenAI Blog
July 17, 2017
We’ve created images that reliably fool neural network classifiers when viewed from varied scales and perspectives. This challenges a claim from last week that self-driving cars would be hard to trick maliciously since they capture images from multiple scales, angles, perspectives, and the like.
Verticals
airesearch
Originally published on OpenAI Blog on 7/17/2017