Two Former Graduate Students Receive the Larry S. Davis Doctoral Dissertation Award

Two Ph.D. graduates advised by faculty in the University of Maryland Institute for Advanced Computer Studies (UMIACS) have been honored for the excellence of their research and scholarship.

Nitin J. Sanket(left) and Soheil Behnezhad (right) are this year’s recipients of the Larry S. Davis Doctoral Dissertation Award, which recognizes outstanding dissertations that convey excellence in technical depth, significance, potential impact and presentation quality.

The award is named for Larry S. Davis, a Distinguished University Professor Emeritus of computer science who served as chair of the department from 1999–2012 and founding director of UMIACS from 1985–1994.

“In my view, Nitin’s thesis is among the best theses that have come out of my lab in the last 15 years,” says Yiannis Aloimonos, a professor of computer science who leads the Perception and Robotics Group (PRG). “Nitin is a very independent, well-organized and highly creative leader with strong technical skills and excellent communication capabilities.”

His dissertation, “Active Vision Based Embodied-AI Design for Nano-UAV Autonomy,” introduces concepts used to develop a novel framework for algorithmic sensorimotor design of multirotor vehicles. In particular, Sanket showcases four methods that achieve activeness on an aerial robot.

His thesis also introduces the RoboBeeHive, a self-navigating aerial drone designed to pollinate flowers on a farm. The drone “hive” houses several much smaller drones and then releases them to pollinate flowers and crops—just like bees.

Using on-board sensing and computation, Sanket has created a new area of research called minimalist cognition. His work—which has been applauded by aerial robotics experts—will enable abilities in robots on a scale never seen before.

“Nitin will be a leading figure in the field in five years’ time,” Aloimonos says, adding that Sanket is adept at working with both hardware and software since he has “one foot in engineering and the other in computer science.”

Sanket is currently a postdoctoral researcher in PRG and an assistant clinical professor in the First-Year Innovation and Research Experience (FIRE) course Autonomous Unmanned Systems.

Behnezhad’s dissertation, “Modern Large-Scale Algorithms for Classical Graph Problems,” studies classical textbook graph problems that have been pondered for decades and proposes modern algorithms that work against huge amounts of data.

In recommending Behnezhad for the award, his adviser Mohammad T. Hajiaghayi noted the dissertation was one of the “deepest and most exciting” that he has read throughout his career.

“Soheil’s dissertation has far-reaching consequences both theoretically and practically,” says Hajiaghayi, the Jack and Rita G. Minker Professor of Computer Science. “It advances our understanding of large-scale algorithms on many fronts.”

The thesis identifies several fundamental problems for massive graphs where traditional algorithms are no longer applicable—often because they’re larger than a single machine’s memory—and presents new algorithms including maximum matching, maximal independent set, minimum vertex cover, and graph connectivity.

Behnezhad began his position as Stanford University’s Motwani Postdoctoral Fellow in August, and will join the Khoury College of Computer Sciences at Northeastern University as an assistant professor next fall.

—Story by Melissa Brachfeld