Jonathan Shapiro

Jonathan Shapiro is an Associate Professor in the School of Computer Science, specializing in machine learning, probabilistic modeling, and evolutionary algorithms. His research emphasizes reinforcement learning, active learning for optimization, and hybrid systems. He leads the Machine Learning and Optimization Research Group and serves as Deputy Director of the interdisciplinary CICADA project, spanning Computer Science, Control Engineering, and Mathematics.

• Current Affiliation: School of Computer Science
• Leadership Roles: Head of Machine Learning and Optimization Research Group, Deputy Director of CICADA
• Former Affiliation: Santa Fe Institute (Adaptive Computation, Evolutionary Dynamics)

His research focuses on adaptive systems, including multi-agent reinforcement learning, evolutionary computation in games, and optimization in noisy environments. He also investigates alarm network redundancy reduction and device heterogeneity in indoor positioning systems.

Jonathan's recent research output explores Bayesian networks, alarm systems, and evolutionary algorithms in games. Key themes include probabilistic modeling, reinforcement learning, and intrinsic delay in dynamical systems. His work contributes to the fields of computer science, optimization, and hybrid systems.

He has supervised projects like New generation algorithms using historical data and market and human behavior change to optimize clients' return on investment (2017–2018) and contributed to economic impacts by improving industrial alarm systems' efficiency and safety.