Valmik Prabhu

Berkeley College, University of California, Berkeley

Papers

Total Citations

H-Index

About

Valmik Prabhu is a researcher specializing in nonlinear control theory, reinforcement learning, and cyber-physical systems design. His most influential work centers on bridging classical control techniques with modern machine learning, particularly through his pioneering research on feedback linearization for systems with unknown dynamics. By leveraging model-free policy optimization, Prabhu developed novel frameworks that enable controllers to learn linearizing strategies for nonlinear plants without requiring explicit knowledge of the underlying system model — a significant advance for real-world deployments where precise dynamics are difficult to characterize. This line of research, spanning a 2019 conference paper and a refined 2020 follow-up, has collectively garnered over 54 citations, reflecting strong community interest in data-driven nonlinear control. More recently, Prabhu has expanded his focus toward the automated co-design of cyber-physical systems, exploring iterated optimization techniques to navigate complex, multi-domain design spaces more efficiently. His work sits at a productive intersection of control engineering, machine learning, and systems design, making his research particularly relevant for students and practitioners interested in intelligent, autonomous, and adaptive engineering systems operating in uncertain real-world environments.