Formal Reasoning About Confidence and Automated Verification of Neural Networks
Open Access
- 18 May 2026
- book chapter
- Published by Springer Nature
- p. 480-499
Abstract
In the last decade, a large body of work has emerged on robustness of neural networks, i.e., checking if the decision remains unchanged when the input is slightly perturbed. However, most of these approaches ignore the confidence of a neural network on its output. In this work, we aim to develop a generalized framework for formally reasoning about the confidence along with robustness in neural networks. We propose a simple yet expressive grammar that captures various confidence-based specifications. We develop a novel and unified technique to verify all instances of the grammar in a homogeneous way, viz., by adding a few additional layers to the neural network, which enables the use any state-of-the-art neural network verification tool. We perform an extensive experimental evaluation over a large suite of 8870 benchmarks, where the largest network has 138M parameters, and show that this outperforms ad-hoc encoding approaches by a significant margin.Keywords
All Related Versions
This publication has 21 references indexed in Scilit:
- Marabou 2.0: A Versatile Formal Analyzer of Neural Networks2024
- Verifying Global Two-Safety Properties in Neural Networks with Confidence2024
- CCN+: A neuro-symbolic framework for deep learning with requirementsInternational Journal of Approximate Reasoning, 2024
- Architecturing Binarized Neural Networks for Traffic Sign Recognition2023
- Neural Network Robustness as a Verification Property: A Principled Case Study2022
- DNNV: A Framework for Deep Neural Network Verification2021
- Reluplex: a calculus for reasoning about deep neural networksFormal Methods in System Design, 2021
- Formal Verification of Piece-Wise Linear Feed-Forward Neural Networks2017
- Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networks2017
- Safety Verification of Deep Neural Networks2017