Mesrob I. Ohannessian received his B.Eng. in Computer and Communications Engineering (2002) from the American University of Beirut and his S.M. in Information Technology (2005) and PhD in Electrical Engineering and Computer Science (2012) from MIT. Before joining TTIC, he was a postdoctoral researcher at UC San Diego. During this time, he spent a semester as a visitor at the Simons Institute at UC Berkeley. He also spent two postdoctoral years in France, one at the Microsoft Research - Inria Joint Centre and another at Université Paris-Sud.
His research interests are in machine learning, statistics, information theory, and their applications, particularly to problems marked by data scarcity. His thesis bridged rare events in discrete spaces with the theory behind continuous extremes. He continues to work on this topic as a means to enhance learning algorithms, since good models for rare events can be thought of as additional structure or regularization in otherwise complex problems. He is interested in various applications, including modeling natural language, climate extremes, and rare-yet-impactful societal events. He is also interested in interactions between statistics and computation, such as striking a tradeoff between accuracy and running time, especially using approaches that smartly “scarcify” data.