Safe Flow Q Learning

Train offline safe RL agents using Hamilton-Jacobi reachability principles to learn feasibility-gated policies. Combine reward and safety critics with flow-matching teacher policies, distill to one-step actors, and calibrate safety thresholds via conformal prediction—achieving near-zero constraint violations with 2.5× inference speedup.

adu2021 Updated 7 repo stars

File contents

adu2021/skillxiv commit 25165995ff

Frequently asked questions

npx get-skill add adu2021/safe-flow-q-learning