TY - CPAPER TI - Concept Enforcement and Modularization as Methods for the ISO 26262 Safety Argumentation of Neural Networks AU - Schwalbe, Gesina AU - Schels, Martin PY - 2020 T2 - Proceeding of the 10th European Congress on Embedded Real Time Software and Systems SP - 1-10 CY - Toulouse UR - https://fis.uni-bamberg.de/handle/uniba/47860 LA - eng AB - 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. KW - concept enforcement KW - machine learning KW - neural networks KW - functional safety KW - ISO 26262 KW - goal structuring notation KW - explainable AI ER -