Provably Tightest Linear Approximation for Robustness Verification of Sigmoid-like Neural Networks.

Zhaodi Zhang Yiting Wu Si Liu Jing Liu Min Zhang

Published in: ASE (2022)

Keyphrases

linear approximation
neural network
activation function
basis functions
pattern recognition
back propagation
model checking
artificial neural networks
worst case
policy iteration
machine learning
high dimensional
real valued