Reinforcement Learning

Predictive Safety Curricula

Adaptive allocation of RL training experience using predictions of future safety cost. Evaluated in controlled simulation and physical-system deployment.

63% lower shank-collision incidence on ANYmal-D hardware across three matched training seeds.

FASTRL: Reinforcement Learning in Surgical Digital Twins

Peer-reviewed work on reinforcement-learning benchmarks and assistance policies in surgical digital-twin environments.

Peer-reviewed benchmark and assistance-policy study in surgical digital twins.

Tracing advantage collapse in RL post-training

A measurement plan for locating where useful variation disappears between reward, advantage estimation, and policy updates.