M
Marco Barreno
Researcher at University of California, Berkeley
Publications - 8
Citations - 2379
Marco Barreno is an academic researcher from University of California, Berkeley. The author has contributed to research in topics: Inductive transfer & Instance-based learning. The author has an hindex of 7, co-authored 8 publications receiving 2039 citations. Previous affiliations of Marco Barreno include University of California.
Papers
More filters
Proceedings ArticleDOI
Can machine learning be secure
TL;DR: A taxonomy of different types of attacks on machine learning techniques and systems, a variety of defenses against those attacks, and an analytical model giving a lower bound on attacker's work function are provided.
Journal ArticleDOI
The security of machine learning
TL;DR: A taxonomy identifying and analyzing attacks against machine learning systems is presented, showing how these classes influence the costs for the attacker and defender, and a formal structure defining their interaction is given.
Proceedings Article
Exploiting machine learning to subvert your spam filter
Blaine Nelson,Marco Barreno,Fuching Jack Chi,Anthony D. Joseph,Benjamin I. P. Rubinstein,Udam Saini,Charles Sutton,J. D. Tygar,Kai Xia +8 more
TL;DR: This paper shows how an adversary can exploit statistical machine learning, as used in the SpamBayes spam filter, to render it useless--even if the adversary's access is limited to only 1% of the training messages.
Proceedings ArticleDOI
Selfish caching in distributed systems: a game-theoretic analysis
Byung-Gon Chun,Kamalika Chaudhuri,Hoeteck Wee,Marco Barreno,Christos H. Papadimitriou,John Kubiatowicz +5 more
TL;DR: The existence of pure strategy Nash equilibria is shown, the price of anarchy is investigated, and the game can always implement the social optimum in the best case by giving servers incentive to replicate.
Book ChapterDOI
Misleading Learners: Co-opting Your Spam Filter
Blaine Nelson,Marco Barreno,Fuching Jack Chi,Anthony D. Joseph,Benjamin I. P. Rubinstein,Udam Saini,Charles Sutton,J. D. Tygar,Kai Xia +8 more
TL;DR: It is shown how an adversary can exploit statistical machine learning, as used in the SpamBayes spam filter, to make it useless—even if the adversary's access is limited to only 1% of the spam training messages.