Differential and Explainable Reinforcement Learning for Multi-objective Optimization in Particle Accelerators
Invited Talk Explainable and Differential Reinforcement Learning for Multi-objective Optimization in Particle Accelerators
admin
•
October 27, 2025 11:00 AM — October 29, 2025 11:15 AM
•
1 min read
Harnessing the power of gradient-based simulations for multi-objective optimization in particle accelerators
Differential Reinforcement Learning outperforms Multi-objective TD3, Multi-Objective Bayesian Optimization, Multi-objective Genetic Algorithms on highly complex constrained …
rajput-k.
Harnessing the Power of Gradient Based Simulations for Multi-Objective Optimization in Particle Accelerators
Invited Talk Talk on applying fast optimization and control with differential reinforcement learning leveraging differential simulations.
admin
Multi-objective Reinforcement Learning
Invited Lecture Lecture on Multi-objective Reinforcement Learning as applied to Particle Accelerators at MaLAPA 2025, CERN, Geneva.
admin
Explainable physics-based constraints on reinforcement learning for accelerator controls
By examining the mathematical form of the learned constraint function, we are able to confirm the agent has learned to use the established physics of each environment. In addition, …
colen-j.
SOCT (Scientific Optimization Control Toolkit)
Scientific Optimization Control Toolkit (SOCT) is an open-source Python framework built on OpenAI Gymnasium for designing and deploying reinforcement learning agents for scientific …
•
1 min read