TY - CPAPER TI - Strategies for Safety Goal Decomposition for Neural Networks AU - Schwalbe, Gesina AU - Schels, Martin PY - 2020 PB - Otto-Friedrich-Universität CY - Bamberg DO - 10.20378/irb-47274 UR - https://fis.uni-bamberg.de/handle/uniba/47274 LA - eng AB - Neural networks (NNs) have become a key technology for solving highly complex tasks, and require integration into future safety argumentations. New safety relevant aspects introduced by NN based algorithms are: representativity of test cases, robustness, inner representation and logic, and failure detection for NNs. In this paper, a general argumentation structure for safety cases respecting these four aspects is proposed together with possible sources of evidence. KW - neural network KW - safety argumentation KW - safety goal decomposition ER -