Enhancing Fault Tolerance of Neural Networks for Security-Critical Applications
This work addresses fault tolerance issues in neural networks for security-critical domains like cryptography, but it appears incremental as it builds on existing partial fault tolerance properties.
The paper tackled the problem of biased partial fault tolerance in neural networks for security-critical applications, proposing a revised implementation that significantly enhances fault tolerance, as demonstrated with concrete improvements in AES SBox implementations on software and FPGA.
Neural Networks (NN) have recently emerged as backbone of several sensitive applications like automobile, medical image, security, etc. NNs inherently offer Partial Fault Tolerance (PFT) in their architecture; however, the biased PFT of NNs can lead to severe consequences in applications like cryptography and security critical scenarios. In this paper, we propose a revised implementation which enhances the PFT property of NN significantly with detailed mathematical analysis. We evaluated the performance of revised NN considering both software and FPGA implementation for a cryptographic primitive like AES SBox. The results show that the PFT of NNs can be significantly increased with the proposed methodology.