BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Talk: Attacks\, defenses\, and the puzzle of machine learning: 
 what data from 10 million hosts can teach us about security
DTSTART:20180427T101500
DTEND:20180427T113000
DTSTAMP:20260506T015641Z
UID:877b865aa9c274a97260327d241f9c13026a741d04a3a0dce5a91f1e
CATEGORIES:Conferences - Seminars
DESCRIPTION:By: Tudor Dumitras - University of Maryland\n\nAbstract:\nWhat
  are the odds that you will get hacked tomorrow? To answer this question\,
  we must understand the capabilities of real-world adversaries and what he
 lps them exploit software vulnerabilities and distribute malware around th
 e world. Moreover\, the machine learning techniques that drive the success
  of such inferences in non-adversarial domains\, like computer vision or a
 utonomous driving\, face new challenges in security.\n \nIn this talk I w
 ill discuss my work\, combining machine learning with global-scale measure
 ments\, that has exposed critical security threats and has guided industri
 al practices. First\, I will present the Worldwide Intelligence Network En
 vironment (WINE)\, an analytics platform that has enabled systematic studi
 es of security threats across more than 10 million hosts from around the w
 orld. Second\, I will use WINE as a vehicle for exploring open questions i
 n security\, such as the duration and impact of zero-day attacks and wheth
 er we can use machine learning to predict which vulnerabilities are going 
 to be exploited. I will also describe applications of these techniques in 
 the emerging cyber insurance industry. I will conclude by discussing how t
 hese results have taught us important lessons about the use of machine lea
 rning in the security domain.\n\nBio:\nTudor Dumitraș is an Assistant Pro
 fessor in the Electrical & Computer Engineering Department at the Universi
 ty of Maryland\, College Park. His research focuses on data-driven securit
 y: he studies real-world adversaries empirically\, he builds  machine  l
 earning systems for detecting attacks and predicting security incidents\, 
 and he investigates the security of machine learning in adversarial enviro
 nments. In his previous role at Symantec Research Labs he built the Worldw
 ide Intelligence Network Environment (WINE) - a data analytics platform fo
 r security research. His work on the effectiveness of certificate revocati
 ons in the Web PKI was featured in the Research Highlights of the Communic
 ations of the ACM in 2018\, and his measurement of the duration and preval
 ence of zero-day attacks received an Honorable Mention in the NSA competit
 ion for the Best Scientific Cybersecurity Paper of 2012. He also received 
 the 2011 A. G. Jordan Award from the ECE Department at Carnegie Mellon Uni
 versity\, the 2009 John Vlissides Award from ACM SIGPLAN\, and the Best Pa
 per Award at ASP-DAC'03. Tudor holds a Ph.D. degree from Carnegie Mellon U
 niversity.\n\nMore information
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR