Develops verifiable error bounds for PINN solutions of Lyapunov and HJB PDEs that turn residual bounds into relative error bounds, certified value function bounds, and valid Lyapunov functions.
Physics-informed neural network Lyapunov functions: PDE charac- terization, learning, and verification.Automatica, 175:112193
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.SY 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Verifiable Error Bounds for Physics-Informed Neural Network Solutions of Lyapunov and Hamilton-Jacobi-Bellman Equations
Develops verifiable error bounds for PINN solutions of Lyapunov and HJB PDEs that turn residual bounds into relative error bounds, certified value function bounds, and valid Lyapunov functions.