Verifying Global Neural Network Specifications using Hyperproperties

12 pages•Published: October 23, 2023

Abstract

Current approaches to neural network verification focus on specifications that target small regions around known input data points, such as local robustness. Thus, using these approaches, we can not obtain guarantees for inputs that are not close to known inputs. Yet, it is highly likely that a neural network will encounter such truly unseen inputs during its application. We study global specifications that — when satisfied — provide guarantees for all potential inputs. We introduce a hyperproperty formalism that allows for expressing global specifications such as monotonicity, Lipschitz continuity, global robustness, and dependency fairness. Our formalism enables verifying global specifications using existing neural network verification approaches by leveraging capabilities for verifying general computational graphs. Thereby, we extend the scope of guarantees that can be provided using existing methods. Recent success in verifying specific global specifications shows that attaining strong guarantees for all potential data points is feasible.

Keyphrases: deep learning, hyperproperties, neural network verification, safe machine learning, trustworthy machine learning

In: Nina Narodytska, Guy Amir, Guy Katz and Omri Isac (editors). Proceedings of the 6th Workshop on Formal Methods for ML-Enabled Autonomous Systems, vol 16, pages 71-82.

Links:	https://easychair.org/publications/paper/JF5L
	https://doi.org/10.29007/pvtn