Pablo A. Parrilo

Pablo A. Parrilo is the Joseph F. and Nancy P. Keithley Professor in the Department of Electrical Engineering and Computer Science at MIT. He is also affiliated with the Laboratory for Information and Decision Systems (LIDS). His research focuses on optimization, control systems, and algebraic methods in engineering and computer science. Parrilo's work bridges convex algebraic geometry and computational tools like semidefinite programming, with applications in machine learning, robotics, and systems theory.

Key areas of research include semidefinite relaxations for polynomial optimization, sum of squares (SOS) programming, and their applications in control and robotics. He co-authored the influential book Semidefinite Optimization and Convex Algebraic Geometry, which explores the intersection of optimization and algebraic geometry.

Parrilo's recent articles emphasize acceleration techniques in optimization, shortest-path problems in convex settings, and algorithmic frameworks for reward learning. His work has been supported by NSF grants such as the AF-funded project on algebraic proof systems and convexity. He has advised numerous graduate students and postdocs, many of whom hold academic and industry positions globally.

His software contributions include the SOSTOOLS MATLAB toolbox for SOS programming, widely used in academia and industry for solving polynomial optimization problems. Parrilo teaches advanced courses on algebraic techniques, optimization methods, and machine learning at MIT.