![[2511.07293] Formal Reasoning About Confidence and Automated Verification of Neural Networks](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2511.07293] Formal Reasoning About Confidence and Automated Verification of Neural Networks
Summary
This paper presents a framework for formal reasoning about the confidence and robustness of neural networks, proposing a unified technique for verification that significantly outperforms existing methods.
Why It Matters
As neural networks are increasingly used in critical applications, understanding their confidence levels alongside robustness is essential for ensuring reliability. This research contributes to the field of AI safety by providing a systematic approach to verifying neural networks, which could enhance trust in AI systems.
Key Takeaways
- Introduces a grammar for confidence-based specifications in neural networks.
- Presents a novel technique for verifying neural networks that integrates confidence and robustness.
- Demonstrates significant performance improvements over traditional verification methods through extensive benchmarking.
Computer Science > Logic in Computer Science arXiv:2511.07293 (cs) [Submitted on 10 Nov 2025 (v1), last revised 14 Feb 2026 (this version, v2)] Title:Formal Reasoning About Confidence and Automated Verification of Neural Networks Authors:Mohammad Afzal, S. Akshay, Blaise Genest, Ashutosh Gupta View a PDF of the paper titled Formal Reasoning About Confidence and Automated Verification of Neural Networks, by Mohammad Afzal and 3 other authors View PDF 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. Subjects: Logic in Computer Science (cs.LO); Artificial Intelligence (...