Dinesh Jayaraman

Dinesh Jayaraman is an Assistant Professor in the Department of Computer and Information Science at the University of Pennsylvania, where he leads the Perception, Action, & Learning Group. His research investigates the rich bidirectional relationship between perception and action: how intelligent agents can use visual and tactile sensing to guide effective actions, and how active interaction with the physical world can in turn produce better, more structured perceptual representations. Jayaraman’s work spans visual control, model-based reinforcement learning, self-supervised representation learning, active perception, visuo-tactile manipulation, causal inference, semantic visual understanding, and zero-shot generalization. His group aims to develop algorithms that unify perception and control to enable robust, affordable, general-purpose robotic systems capable of operating in unstructured environments. Prior to joining Penn, Jayaraman was a visiting research scientist at Facebook AI Research in Menlo Park and a postdoctoral scholar at UC Berkeley. He received his Ph.D. from UT Austin and his Bachelor’s degree from IIT Madras. His research integrates foundational machine learning with robot embodiment, with the goal of building agents that learn effectively through perception–action loops.