BEGIN:VCALENDAR
VERSION:2.0
PRODID:https://github.com/derhansen/sf_event_mgt
METHOD:PUBLISH
BEGIN:VEVENT
UID:5763-409@gi.de
CLASS: PUBLIC
SUMMARY:ATHENE Distinguished Lecture Series
DESCRIPTION:Konrad Rieck is a professor at TU Berlin, where he leads the Ch
 air of Machine Learning and Security as part of the Berlin Institute for th
 e Foundations of Learning and Data. Previously, he held academic positions 
 at TU Braunschweig, the University of Göttingen, and Fraunhofer Institute F
 IRST. His research focuses on the intersection of computer security and mac
 hine learning. He has published over 100 papers in this area and serves on 
 the PCs of the top security conferences (system security circus). He has be
 en awarded the CAST/GI Dissertation Award, a Google Faculty Award, and an E
 RC Consolidator Grant.\n\nWhen Papers Choose their Reviewers: Adversarial M
 achine Learning in Conference Management Systems Abstract\n\n\n\nThe number
  of papers submitted to scientific conferences is steadily rising in many d
 isciplines. To handle this growth, systems for automatic paper-reviewer ass
 ignments are increasingly used during the reviewing process. These systems 
 employ statistical topic models to characterize the papers' content and aut
 omate their assignment to reviewers. In this talk, we invesitgate the secur
 ity of this automation and introduce a new attack that modifies a given pap
 er so that it selects its own reviewers. Our attack is based on a novel opt
 imization strategy that fools the topic model with unobtrusive changes to t
 he paper's content. In an empirical evaluation with a (simulated) conferenc
 e, our attack successfully selects and removes reviewers, while the tampere
 d papers remain plausible and often indistinguishable from innocuous submis
 sions.
LOCATION:Fraunhofer SIT
DTSTAMP:20230710T105335Z
DTSTART:20230912T120000Z
DTEND:20230912T130000Z
END:VEVENT
END:VCALENDAR