ECE/RISCS Seminar with Aryeh Kontorovich

Metric Anomaly Detection Via Asymmetric Risk Minimization Faculty Host: Ari Trachtenberg Abstract: Professor Kontorovich and his research team propose what appears to be the first anomaly detection framework that learns from positive examples only and is sensitive to substantial differences in the presentation and penalization of normal vs. anomalous points. Their framework introduces a novel type of asymmetry between how false alarms (misclassfications of a normal instance as an anomaly) and missed anomalies (misclassfications of an anomaly as normal) are penalized: whereas each false alarm incurs a unit cost, their model assumes that a high global cost is incurred if one or more anomalies are missed. Kontorovich will define a few natural notions of risk along with efficient minimization algorithms. His team’s framework is applicable to any metric space with a finite doubling dimension. They make minimalistic assumptions that naturally generalize notions such as margin in Euclidean spaces. Kontorovich will provide a theoretical analysis of the risk and show that under mild conditions, their classifier is asymptotically consistent. The learning algorithms they propose are computationally and statistically efficient and admit a further tradeoff between running time and precision. Some experimental results on real-world data will be provided. Joint work with Danny Hendler and Eitan Menahem About the Speaker: Aryeh (Leonid) Kontorovich received his undergraduate degree in mathematics with a certificate in applied mathematics from Princeton University in 2001. His M.Sc. and Ph.D. are from Carnegie Mellon University, where he graduated in 2007. After a postdoctoral fellowship at the Weizmann Institute of Science, he joined the Computer Science department at Ben-Gurion University of the Negev in 2009 as an assistant professor; this is his current position. His research interests are mainly in machine learning, with a focus on probability, statistics, and automata theory.