A neural hybrid automaton surrogate is learned from data to enable gradient-based optimization that finds safety violations in hybrid systems more efficiently than prior tools, with final verification on the original system.
Neu- ral ordinary differential equations
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
-
Learning Neural Hybrid Surrogates for Gradient-Based Falsification
A neural hybrid automaton surrogate is learned from data to enable gradient-based optimization that finds safety violations in hybrid systems more efficiently than prior tools, with final verification on the original system.