Skip to Main content Skip to Navigation
New interface
Conference papers

Concept Enforcement and Modularization as Methods for the ISO 26262 Safety Argumentation of Neural Networks

Abstract : Neural networks (NN) are prone to systematic faults which are hard to detect using the methods recommended by the ISO 26262 automotive functional safety standard. In this paper we propose a unified approach to two methods for NN safety argumentation: Assignment of human interpretable concepts to the internal representation of NNs to enable modularization and formal verification. Feasibility of the required concept embedding analysis is demonstrated in a minimal example and important aspects for generalization are investigated. The contribution of the methods is derived from a proposed generic argumentation structure for a NN model safety case.
Complete list of metadata

Cited literature [28 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02442796
Contributor : Gesina Schwalbe Connect in order to contact the contributor
Submitted on : Thursday, January 16, 2020 - 5:03:50 PM
Last modification on : Friday, February 14, 2020 - 9:18:02 AM
Long-term archiving on: : Friday, April 17, 2020 - 7:56:25 PM

File

ERTS2020_paper_10.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02442796, version 1

Collections

Citation

Gesina Schwalbe, Martin Schels. Concept Enforcement and Modularization as Methods for the ISO 26262 Safety Argumentation of Neural Networks. 10th European Congress on Embedded Real Time Software and Systems (ERTS 2020), Jan 2020, Toulouse, France. ⟨hal-02442796⟩

Share

Metrics

Record views

380

Files downloads

233