# Safely Entering the Deep: A Review of Verification and Validation for Machine Learning and a Challenge Elicitation in the Automotive Industry

## TL;DR

This paper reviews current verification and validation methods for safety-critical machine learning systems in automotive applications, highlighting challenges and proposing knowledge transfer and systems-based safety approaches.

## Contribution

It provides a comprehensive review of verification and validation techniques for DNNs in automotive safety and reports expert insights on adapting safety standards.

## Key findings

- ISO 26262 conflicts with DNN characteristics
- Knowledge transfer from aerospace enhances automotive safety
- Safety cage architectures improve DNN safety assurance

## Abstract

Deep Neural Networks (DNN) will emerge as a cornerstone in automotive software engineering. However, developing systems with DNNs introduces novel challenges for safety assessments. This paper reviews the state-of-the-art in verification and validation of safety-critical systems that rely on machine learning. Furthermore, we report from a workshop series on DNNs for perception with automotive experts in Sweden, confirming that ISO 26262 largely contravenes the nature of DNNs. We recommend aerospace-to-automotive knowledge transfer and systems-based safety approaches, e.g., safety cage architectures and simulated system test cases.