Towards a Safety Case for Hardware Fault Tolerance in Convolutional Neural Networks Using Activation Range Supervision

TL;DR

This paper proposes a safety framework for CNNs in safety-critical systems, using activation range supervision to detect and mitigate hardware soft errors, demonstrated on vehicle classification with ResNet-50.

Contribution

It introduces activation range supervision as a reliable fault detection method and explores novel range restriction techniques for CNN safety in hardware fault scenarios.

Findings

Activation range supervision effectively detects hardware soft errors.

Non-uniform range restriction reduces silent data corruptions.

Demonstrated safety benefits in vehicle classification with ResNet-50.

Abstract

Convolutional neural networks (CNNs) have become an established part of numerous safety-critical computer vision applications, including human robot interactions and automated driving. Real-world implementations will need to guarantee their robustness against hardware soft errors corrupting the underlying platform memory. Based on the previously observed efficacy of activation clipping techniques, we build a prototypical safety case for classifier CNNs by demonstrating that range supervision represents a highly reliable fault detector and mitigator with respect to relevant bit flips, adopting an eight-exponent floating point data representation. We further explore novel, non-uniform range restriction methods that effectively suppress the probability of silent data corruptions and uncorrectable errors. As a safety-relevant end-to-end use case, we showcase the benefit of our approach in a vehicle classification scenario, using ResNet-50 and the traffic camera data set MIOVision. The quantitative evidence provided in this work can be leveraged to inspire further and possibly more complex CNN safety arguments.