Incorrect by Construction: Fine Tuning Neural Networks for Guaranteed Performance on Finite Sets of Examples

Ivan Papusha
Ivan Papusha
Rosa Wu
Rosa Wu
Joshua Brulé
Joshua Brulé
Daniel Genin
Daniel Genin
Cited by: 0|Bibtex|Views0|Links

Abstract:

There is great interest in using formal methods to guarantee the reliability of deep neural networks. However, these techniques may also be used to implant carefully selected input-output pairs. We present initial results on a novel technique for using SMT solvers to fine tune the weights of a ReLU neural network to guarantee outcomes o...More

Code:

Data:

Your rating :
0

 

Tags
Comments