Paul Ruvolo, Ph.D.
EducationPh.D., Computer Science and Engineering, University of California San Diego
M.S., Computer Science and Engineering, University of California San Diego
B.S., Computer Science, Harvey Mudd College
Select Courses TaughtModeling and Simulation of the Physical World
AwardsNSF IGERT Fellowship for studying "Learning and Vision in Humans and Machines"
Dr. Ruvolo's goal is to develop machines capable of emulating the incredible perceptual and motor abilities of humans. Historically, machines, e.g. software systems or robots, have had to be meticulously preprogrammed to solve a given sensorimotor task. While this approach has been successful for sufficiently constrained tasks, this approach has proven to be brittle when presented with the unconstrained complexity inherent in daily life. In contrast, Dr. Ruvolo's approach is to create systems capable of learning and adapting to novel situations. This learning takes two forms: (1) directly imitating human sensory and motor behavior and (2) autonomously adapting through experience. In order to create machines capable of learning through both imitation and experience, he leverages mathematical tools from fields such as Bayesian statistics, numerical optimization, and linear algebra. Using this approach he has tackled problems in computer vision (making software that can make sense of the visual world), computer audition (making software that can make sense of the auditory world), and optimal control (making software capable of optimally solving sequential decision-making tasks). His research focus at Olin is using this learning approach to create assistive technologies for people with sensory and motor impairments.