ReachNN abstracts feedforward neural networks with Bernstein polynomials and provides error bounds to compute reachable sets for verifying neural-network controlled systems with general Lipschitz-continuous activation functions.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.SY 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ReachNN: Reachability Analysis of Neural-Network Controlled Systems
ReachNN abstracts feedforward neural networks with Bernstein polynomials and provides error bounds to compute reachable sets for verifying neural-network controlled systems with general Lipschitz-continuous activation functions.