CT-BaB integrates branch-and-bound during training to tighten certified Lyapunov bounds, yielding neural controllers with 164X larger verifiable ROA and 11X faster verification than CEGIS on a 2D quadrotor.
Neural certificates for safe control policies
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
Develops sufficient conditions and sum-of-squares verification algorithms for union of control barrier functions under two switching strategies to ensure finite switches and forward invariance.
citing papers explorer
-
Certified Training with Branch-and-Bound for Lyapunov-stable Neural Control
CT-BaB integrates branch-and-bound during training to tighten certified Lyapunov bounds, yielding neural controllers with 164X larger verifiable ROA and 11X faster verification than CEGIS on a 2D quadrotor.
-
Verification Framework for the Union of Control Barrier Functions
Develops sufficient conditions and sum-of-squares verification algorithms for union of control barrier functions under two switching strategies to ensure finite switches and forward invariance.