News for hackers. Stuff that matters.
Re:Slight Street Sign Modifications Can Fool Machine Learnin... by moyix on Saturday August 05, 2017 @11:06AM (#14936051)
There was a paper that claimed those sorts of rotations and movements can help alleviate adversarial examples:

https://arxiv.org/abs/1707.03501

However, OpenAI quickly refuted it by creating adversarial examples that continue to fool the classifier even when rotated, scaled, etc:

https://blog.openai.com/robust-adversarial-inputs/

So it looks like there's no "easy" way out here. Multiple types of sensors may help, but it seems likely that it will still be possible to construct examples that fool network over all sensor inputs at once.

Ian Goodfellow and Nicolas Papernot have a good blog on machine learning security issues. One relevant post on why this is such a hard problem:

http://www.cleverhans.io/security/privacy/ml/2017/02/15/why-...

Re:Slight Street Sign Modifications Can Fool Machine Learnin... by pishpash on Saturday August 05, 2017 @11:15AM (#14936093)
Is it a big surprise that you can construct adversarial examples for algorithms? Don't humans have the same class of problems with optical illusions? And those are not even adversarial, just confusing.

If we constructed truly adversarial examples for human neurology, I bet they would be equally insane.

Re:Slight Street Sign Modifications Can Fool Machine Learnin... by blauditore on Saturday August 05, 2017 @11:34AM (#14936201)
In some sense, yes, optical illusions are simlar to such "adversarial examples". But if you think about it, any kind of image is somewhat delusive, since we perceive it as whatever object it depicts while actually staring at a piece of paper with some ink on it.

Also, adversarial in this case seems to refer to images perceived differently by machines than by humans, so it's not really possible to create such ones for humans.

Re:Slight Street Sign Modifications Can Fool Machine Learnin... by kthejoker2 on Sunday August 06, 2017 @01:15AM (#14939830)
No, adversarial simply means deliberately trying to engineer false positives and negatives. This can be done against humans, machines, ants, trees, viruses ...
Re:Slight Street Sign Modifications Can Fool Machine Learnin... by blauditore on Tuesday August 08, 2017 @04:00AM (#14955523)
What is a false positive or false negative in this case? The "ground truth" here is what human perceive.
Re:Slight Street Sign Modifications Can Fool Machine Learnin... by kthejoker2 on Wednesday August 09, 2017 @08:56PM (#14975318)
Optical illusion?
Re:Slight Street Sign Modifications Can Fool Machine Learnin... by robotresearcher on Saturday August 05, 2017 @04:35PM (#14937767)
Camouflage is an adversarial example.
Re:Slight Street Sign Modifications Can Fool Machine Learnin... by bluewolf_sk on Saturday August 05, 2017 @04:01PM (#14937591)
Wait... the dress was blue and black?
Re:Slight Street Sign Modifications Can Fool Machine Learnin... by thedevil on Saturday August 05, 2017 @11:21AM (#14936127)
Thank you for your awesome response. It probably would have taken me hours of googling and reading to find these exact posts. A mere upvote is not enough of a thank you for such a quality reply.