Towards Dependability Metrics for Neural Networks
This work addresses the urgent need for standardized dependability metrics in safety-critical applications like automated driving, though it appears incremental as it builds on existing concepts without introducing a new paradigm.
The paper tackles the challenge of establishing safety engineering practices for neural networks in automated driving by proposing a set of efficiently computable metrics to measure dependability attributes such as robustness, interpretability, completeness, and correctness.
Artificial neural networks (NN) are instrumental in realizing highly-automated driving functionality. An overarching challenge is to identify best safety engineering practices for NN and other learning-enabled components. In particular, there is an urgent need for an adequate set of metrics for measuring all-important NN dependability attributes. We address this challenge by proposing a number of NN-specific and efficiently computable metrics for measuring NN dependability attributes including robustness, interpretability, completeness, and correctness.