[Pycon] [new paper] "Amit Kushwaha" - Security and Privacy in Machine Learning: Adversarial machine learning

[info a pycon.it]

Title: Security and Privacy in Machine Learning: Adversarial machine learning
Duration: 45 (includes Q&A)
Q&A Session: 15
Language: en
Type: Talk

Abstract: Security and Privacy issues need no introduction. But how exactly this is affecting the field of Machine Learning? This is what this talk will cover. We first expose the attack surface of systems deploying machine learning. We then describe how an attacker may force models to make wrong predictions with very little information about the victim. We demonstrate that these attacks are practical against existing machine learning as a service platform. Finally, we will discuss current researches to defend them from such attacks.


Tags: [u'privacy', u'security', u'Artificial Intelligence']