Speaker
Sabrina Appel
(GSI)
Description
Automation & Optimization with Classical Optimizer towards RL at GSI with Geoff
Geoff is an open-source optimization framework developed at CERN and further maintained in collaboration with GSI as part of the EURO-LABS project. It provides a standardized interface to optimization problems, enabling seamless switching between classical optimizers and reinforcement learning approaches. In addition, utility functions and tools within Geoff help accelerate the development of optimization workflows. Using the example of beam injection into SIS18 at GSI, we illustrate how optimization can evolve from classical optimizer to reinforcement learning.
Primary authors
Dr
Penelope Madysa
(GSI)
Sabrina Appel
(GSI)
