Материал түпнұсқа тілінде: English.
Robust adversarial inputs
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.
2017 ж. 17 шілде · Дереккөз: OpenAI news
Толық мәтін дереккөзде қолжетімді.