Mohit Iyyer and Srijan Kumar awarded Larry S. Davis Doctoral Dissertation Awards for 2017

Mohit Iyyer and Srijan Kumar have been named the 2017 recipients of the Larry S. Davis Doctoral Dissertation Award. This dissertation award recognizes the two most outstanding dissertations completed over the past year in the department.  Iyyer’s dissertation is entitled Beyond Sentential Language Understanding with Deep Learning; he was advised by Jordan Boyd-Graber and Hal Daumé III. Kumar’s dissertation is entitled Characterization and Detection of Malicious Behavior on the Web, and he was advised by V.S. Subrahmanian.

A committee of professorial faculty solicit nominations for this award and make final decisions. “All the nominations we received were very strong,” said Professor Jeff Foster, Associate Chair of Graduate Education, “so it was a difficult decision.”

“I'm very pleased with the two winners of this year's competition.  I was on the selection committee and was impressed by the technical contributions of all the submissions.  I think they reflect the overall very high quality of our Ph. D. program,” said Professor Larry Davis, who also currently serves as Interim Chair of the department.

Iyyer areas of expertise include Artificial Intelligence and Machine Learning. While explaining the work of his dissertation, he said, “For a computer to understand human language, it has to make sense of not only words and sentences but also discourse-level units such as paragraphs and chapters.”  For the work of his dissertation he said, “I present a diverse array of language-based tasks at the discourse level (e.g., conversational question answering, fictional relationship modeling in novels, and narrative understanding in comic books). I design new deep learning architectures, which construct representations of language directly from text without relying on user-specified features that can solve these tasks by integrating both immediate and high-level context. “

He currently works as a postdoctoral researcher at the Allen Institute for Artificial Intelligence in Seattle, Washington. In Fall Semester 2018, Iyyer will join University of Massachusetts Amherst as an Assistant Professor of Computer Science.

For his dissertation, Kumar explained that he focused on characterizing and detection of malicious behavior on the web and social media, such as trolls, vandals, false information, fake reviews, and sockpuppetry. “These malicious actors reduce the safety, usability, and reliability of web platforms. Studying their behavior is essential in developing early detection methods to catch them before they can do much harm,” he said.  He further explained that “I comprehensively characterize various types of malicious behavior across several different platforms, and develop graph based and feature based algorithms to efficiently detect malicious from benign online behavior.”

Currently, Kumar is a postdoctoral researcher at Stanford University working on network analysis and machine learning with Professor Jure Leskovec.

“Srijan's work represents a significant effort to help create a safer web and a safer social media environment for all of. His work will likely have a lasting impact on those who use platforms such as Wikipedia, Amazon, Flipkart, and Bitcoin - which includes hundreds of millions of people,” said Professor V.S. Subrahmanian, who now serves as Dartmouth College Distinguished Professor in Cybersecurity, Technology, and Society.

Iyyer’s and Kumar's research will no doubt have a significant impact on the future of artificial intelligence, machine learning, and safety in computing.

***
About the Award

Larry S. Davis is Interim Chair of the Department of Computer Science, and he was Chair of the Department of Computer Science from 1999–2012 and Director of the University of Maryland Institute for Advanced Computer Studies from 1985–1994. He was named a Fellow of the IEEE in 1997. Prof. Davis is known for his research in computer vision and high performance computing. He has published over 100 papers in journals and 200 conference papers and has supervised over 35 Ph.D. students. He and his students have developed foundational methods for detection and tracking of people and vehicles in video, representation and recognition of human movements and activities, and mixed AI/signal processing models for event modeling and recognition.

Eligibility

Each nominated dissertation must have been defended between October of the previous year and September of the current year (inclusive). Eligible dissertations that were defended the previous September, but were too late to be nominated before, may also be considered in exceptional circumstances. Each faculty member may nominate at most one dissertation.

Selection

Dissertations will be evaluated based on technical depth, significance, potential impact, and presentation quality. The selection committee will consist of three faculty members selected by the department.  Up to two awards may be granted each year, and up to one honorable mention.

Award

The Larry S. Davis Doctoral Dissertation Award includes a prize of $500. The selected dissertation(s) will normally also be nominated for the ACM Doctoral Dissertation Award by the department.