4th collaboration workshop on Reinforcement Learning for Autonomous Accelerators (RL4AA'26)

Session

Keynote

31 Mar 2026, 09:15

Theatre 2, Teaching Hub 502 (University of Liverpool)

Conveners

Keynote

• Andrea Santamaria Garcia (University of Liverpool and Cockcroft Institute)

31. Accelerating Reinforcement Learning with Off-Policy Data: Promises, Pitfalls, and Future Directions

samuele Tosatto (Universität Innsbruck)

31/03/2026, 09:15

Talk

Reinforcement learning is a promising technique for solving complex control problems in real-world physical systems, such as robotics, plasma stabilization, and particle accelerators. However, RL is often data-hungry, and its classic on-policy formulation is often inefficient, as it disallows data reuse, and unsafe, as it requires the agent to interact with the environment from...