analiza aiqbench
NNV3 represents a significant maturation in formal verification by extending its reachability framework to graph neural networks, 3D volumetric data, and probabilistic conformal inference. These updates address critical gaps in verifying complex AI systems used in cyber-physical domains like power grids and medical imaging, where deterministic methods often fail. The inclusion of fairness certification tools further strengthens the toolkit for ensuring robustness against weight perturbations and counterfactual biases.
NNV3: Expanding Neural Network Verification to New Architectures and Domains
arXiv:2609.30050v1 Announce Type: new Abstract: We present NNV3, the latest version of the Neural Network Verification (NNV) tool, a MATLAB framework for formal verification of deep learning models and learning-enabled cyber-physical systems. Building on the set-based reachability foundation of NNV 1.0 (FFNNs, CNNs, NNCS) and NNV 2.0 (RNNs, SSNNs, neural ODEs), NNV3 introduces new members of the Star-set family: ModelStar for verifying networks under weight perturbation, VolumeStar for video and 3D volumetric inputs, and GraphStar for graph neural networks. A conformal-inference-based probabilistic reachability mode complements sound analysis for problems where deterministic verification is intractable, while FairNNV certifies counterfactual and individual fairness properties over continuous input regions. NNV3 introduces new benchmarks for malware detection, graph-based power-system models, medical imaging, variable-length time series data, and action recognition. NNV3 also incorporates tutorials and developer guides through a unified documentation site. This paper details these major updates, demonstrating NNV's maturation into a comprehensive, robust, and accessible verification tool for a diverse range of AI systems.
Authors:Anne M. Tumlin, Samuel Sasaki, Ben Wooding, Diego Manzanas Lopez, Muhammad Usama Zubair, Navid Hashemi, Hongchao Zhang, Waseem Abbas, Ipek Oguz, Meiyi Ma, Taylor T. Johnson
Abstract:We present NNV3, the latest version of the Neural Network Verification (NNV) tool, a MATLAB framework for formal verification of deep learning models and learning-enabled cyber-physical systems. Building on the set-based reachability foundation of NNV 1.0 (FFNNs, CNNs, NNCS) and NNV 2.0 (RNNs, SSNNs, neural ODEs), NNV3 introduces new members of the Star-set family: ModelStar for verifying networks under weight perturbation, VolumeStar for video and 3D volumetric inputs, and GraphStar for graph neural networks. A conformal-inference-based probabilistic reachability mode complements sound analysis for problems where deterministic verification is intractable, while FairNNV certifies counterfactual and individual fairness properties over continuous input regions. NNV3 introduces new benchmarks for malware detection, graph-based power-system models, medical imaging, variable-length time series data, and action recognition. NNV3 also incorporates tutorials and developer guides through a unified documentation site. This paper details these major updates, demonstrating NNV's maturation into a comprehensive, robust, and accessible verification tool for a diverse range of AI systems.
Artificial Intelligence (cs.AI)
arXiv:2609.30050 [cs.AI]
(or arXiv:2609.30050v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.30050 arXiv-issued DOI via DataCite (pending registration)
Text integral sub licența deschisă a publicației · 200 cuvinte
Subiecte conexe
News
Quantum Computing Report · 25 iul. 2026
NVIDIA GTC Taipei at COMPUTEX: Live Updates on What’s Next in AI
NVIDIA · 21 mai 2026
Finding Success in Industry as a Chip Designer
IEEE Spectrum · 28 mai 2026
What's Continual Learning, and Why Might We Expect To See It In Advanced LLM Agents?
LessWrong · 12 iun. 2026