Nathan Ratliff recently completed Ph.D. work at Carnegie Mellon University’s Robotics Institute under Professor J. Andrew Bagnell. His primary area of research during this period was been imitation learning.

Early in his graduate studies, Nathan, with Professor Bagnell and Dr. Martin Zinkevich, reduced imitation learning (for the case where optimal planners exist) to maximum margin structured classification, and developed online, batch, and functional subgradient methods for efficiently solving the optimization problem that resulted. This framework is known as maximum margin planning (MMP), and their class of linear and nonlinear gradient-based approaches to solving the optimization problem that results is known as LEArning to seaRCH (LEARCH).

Applications of this framework include footstep prediction, grasp prediction, heuristic learning, overhead navigation, LADAR classification, and optical character recognition. His thesis work additionally studies generalizing these ideas to high-dimensional configuration spaces where optimal planning is intractable. Nathan is now at the Toyota Technological Institute (TTI) in Chicago. He collaborates closely with the robotics efforts at the University of Washington and at Intel Research in Seattle.

Publications > http://www.nathanratliff.com/robotlearning/publications.html