A neuron behavior consistency loss reduces the number of unstable neurons and speeds up formal verification of trained networks across MNIST, Fashion-MNIST, and CIFAR-10.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Training Verification-Friendly Neural Networks via Neuron Behavior Consistency
A neuron behavior consistency loss reduces the number of unstable neurons and speeds up formal verification of trained networks across MNIST, Fashion-MNIST, and CIFAR-10.