Formal Monotony Analysis of Neural Networks with Mixed Inputs: An Asset for Certification - Advancing Rigorous Software and System Engineering Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Formal Monotony Analysis of Neural Networks with Mixed Inputs: An Asset for Certification

Résumé

The use of ML technology to design safety-critical systems requires a complete understanding of the neural network’s properties. Among the relevant properties in an industrial context, the verification of partial monotony may become mandatory. This paper proposes a method to evaluate the monotony property using a Mixed Integer Linear Programming (MILP) solver. Contrary to the existing literature, this monotony analysis provides a lower and upper bound of the space volume where the property does not hold, that we denote “Non-Monotonic Space Coverage”. This work has several advantages: (i) our formulation of the monotony property works on discrete inputs, (ii) the iterative nature of our algorithm allows for refining the analysis as needed, and (iii) from an industrial point of view, the results of this evaluation are valuable to the aeronautical domain where it can support the certification demonstration. We applied this method to an avionic case study (braking distance estimation using a neural network) where the verification of the monotony property is of paramount interest from a safety perspective.
Fichier principal
Vignette du fichier
Formal Monotony Analysis of Neural Networks with Mixed Inputs_An Asset for Certification.pdf (1.6 Mo) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte

Dates et versions

hal-03855271 , version 1 (18-11-2022)

Identifiants

Citer

Guillaume Vidot, Mélanie Ducoffe, Christophe Gabreau, Iulian Ober, Ileana Ober. Formal Monotony Analysis of Neural Networks with Mixed Inputs: An Asset for Certification. 27th International Conference on Formal Methods for Industrial Critical Systems (FMICS 2022), Sep 2022, Warsaw, Poland. pp.15-31, ⟨10.1007/978-3-031-15008-1_3⟩. ⟨hal-03855271⟩
102 Consultations
82 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More