Rel Guzman
Papers
2
Total Citations
5
H-Index
2
About
Rel Guzman is an emerging researcher working at the intersection of Bayesian optimization and model predictive control (MPC), with a focus on developing intelligent, adaptive control systems capable of handling real-world uncertainty. Their work addresses a fundamental challenge in control engineering: how to reliably tune and operate MPC systems when the underlying models of physical dynamics are imperfect or stochastic. Guzman's most notable contributions include the development of Bayesian optimization frameworks specifically tailored for MPC hyper-parameter tuning under uncertainty. Their 2022 paper introduced an adaptive approach that jointly estimates probability distributions over model parameters using performance-based rewards, while their earlier 2020 work explored heteroscedastic Bayesian optimization to handle noise that varies across the input space — a critical consideration in stochastic control environments. Together, these works represent a cohesive research agenda aimed at making MPC more robust and self-tuning in practical applications. Though early in their citation trajectory — with 3 and 2 citations respectively — Guzman's research tackles problems of growing importance as autonomous systems and data-driven control become increasingly prevalent. Students interested in probabilistic machine learning applied to control theory will find Guzman's work a valuable and technically rigorous reference point.
Research Focus
Key Achievements
Top Papers
- 1
- 2