Eric Mazumdar

Papers

Total Citations

H-Index

About

Eric Mazumdar is a researcher working at the intersection of reinforcement learning, control theory, and optimization. His work is particularly distinguished by its focus on bridging the gap between classical nonlinear control techniques and modern machine learning methods. Most notably, Mazumdar has pioneered approaches that leverage model-free policy optimization to learn feedback linearizing controllers for physical systems with unknown dynamics — a significant contribution that enables robust control design without requiring explicit knowledge of the underlying plant model. His 2020 paper on feedback linearization for uncertain systems has garnered 36 citations, with a closely related 2019 predecessor accumulating an additional 18 citations, reflecting a sustained and growing interest in this line of work from both the control and machine learning communities. By adapting the classical feedback linearization framework — traditionally reliant on precise system models — to accommodate real-world uncertainty through reinforcement learning, Mazumdar has opened new pathways for deploying intelligent controllers in complex, poorly characterized environments. His research offers valuable tools for students and practitioners seeking to apply learning-based methods to challenging nonlinear control problems with practical engineering relevance.