Re:A Guide to Synthesizing Adversarial Examples
by yters on Tuesday July 25, 2017 @01:50PM
(#14850096)
If it is so easy to fool deep learning, why is it so hyped? Seems a great security risk.
[ Parent
| on HN ]
Re:A Guide to Synthesizing Adversarial Examples
by suryabhupa on Tuesday July 25, 2017 @04:19PM
(#14851488)
Many machine learning and reinforcement learning models are susceptible to adversarial attacks; it's not unique to deep learning. However, because so many systems that are currently deployed in applications use deep learning, it's under particular scrutiny.
Re:A Guide to Synthesizing Adversarial Examples
by yters on Wednesday July 26, 2017 @12:08AM
(#14853600)
Then it seems the hype of machine learning is not well founded. Machine learning in general is a big risk if it so easily fooled.
Re:A Guide to Synthesizing Adversarial Examples
by skykooler on Tuesday July 25, 2017 @03:35PM
(#14851143)
It's hyped because it can guess the right answer "most of the time". As long as it's working with a "random" input (not someone intentionally trying to fool it) it works well; and this is all most people care about.