Strategies for Improving the Error Robustness of Convolutional Neural Networks
Authors:
- 1 December 2022
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 874-883
Abstract
The error robustness of Convolutional Neural Networks (CNNs) is an important attribute requiring attention due to their growing application in safety-critical domains such as autonomous driving and medical devices. Hardware errors affecting the execution of such models may lead to system failures and, therefore, fault tolerance techniques are necessary to improve dependability. This paper proposes an approach to improve the robustness of CNNs and experimentally compares it with three other existing techniques. Fault injection is used to emulate hardware faults affecting CNNs targeting four distinct datasets. Results indicate that the ranger technique globally provides the best robustness closely followed by the stimulated training technique, although the former provides much lower temporal overhead than the latter. Architectural redundancy and dropout provide varying results. In all cases, caution through final evaluation of any CNN is required, because there are corner cases in which the robustness decreases, contrary to the intended outcome.Keywords
Funding Information
- Foundation for Science and Technology
This publication has 30 references indexed in Scilit:
- ImageNet classification with deep convolutional neural networksCommunications of the ACM, 2017
- When Neurons Fail2017
- SRAM voltage scaling for energy-efficient convolutional neural networks2017
- A Triple Core Lock-Step (TCLS) ARM® Cortex®-R5 Processor for Safety-Critical and Ultra-Reliable Applications2016
- Evaluation and Mitigation of Radiation-Induced Soft Errors in Graphics Processing UnitsIEEE Transactions on Computers, 2015
- The German Traffic Sign Recognition Benchmark: A multi-class classification competition2011
- Soft Errors in Advanced Computer SystemsIEEE Design & Test of Computers, 2005
- Comparison of physical and software-implemented fault injection techniquesIEEE Transactions on Computers, 2003
- A modified learning algorithm for improving the fault tolerance of BP networks2002
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998