A hierarchical reinforcement learning and predictive control framework for automated multi-variable anesthesia

Short summary

A hierarchical Reinforcement Learning (RL) agent dynamically tunes a Generalized Predictive Control (GPC) cost function, achieving 50% faster anesthesia induction (median 228 vs. 452 seconds) and maintaining safety across 44 virtual patients compared to a Genetic Algorithm (GA)-optimized GPC.

AI-generated from the title and abstract; the full text is not read.