Jorge Mendez-Mendez

Jorge Mendez-Mendez is an Assistant Professor in the Department of Electrical and Computer Engineering at Stony Brook University. Prior to joining Stony Brook, he was a postdoctoral fellow at MIT CSAIL, and has a formal academic background from the University of Pennsylvania (PhD, 2022; MSE, 2018) and Universidad Simon Bolivar (Bachelor's, 2016).

• PhD and MSE from University of Pennsylvania's GRASP Lab
• Bachelor's in Electronics Engineering from Universidad Simon Bolivar, Venezuela

His research focuses on creating versatile, intelligent, embodied agents that learn through lifelong knowledge accumulation. He explores how agents can leverage compositional and modular structures to simplify complex, continuous learning problems. His work applies to robotics, computer vision, and natural language processing, with key contributions in lifelong machine learning, sequential decision making, robot learning, and reinforcement learning.

Jorge's publications highlight trends in lifelong learning, reinforcement learning, and compositional structures. His recent work includes black-box robustness improvements, bilevel planning, offline compositional reinforcement learning, and continual novelty detection.

Scientific awards include:

• MIT-IBM Distinguished Postdoctoral Fellowship (2022-2024)
• Third Place in Two Sigma Diversity PhD Fellowship (2021)
• Best Paper at Lifelong Learning Workshop, ICML (2020)

He teaches ESE 577: Deep Learning Algorithms and Software.