Differential and Explainable Reinforcement Learning for Multi-objective Optimization in Particle Accelerators

October 27, 2025·
admin
· 0 min read
Image credit: Kishan
Abstract
This talk presents recent advancements in applying explainable and differential reinforcement learning techniques for multi-objective optimization in particle accelerators. We will discuss how differential simulations combined with RL can handle conflicting optimization objectives, and how explainability methods help understand the decision-making process of RL agents.
Date
October 27, 2025 11:00 AM — October 29, 2025 11:15 AM
Location

MIT, Boston, MA

Boston, MA,

events