5 May 2025 in IEEE Transactions on Computers
The complexity of the state-of-the-art devices makes reliability assessments approaches extremely complex and, sometimes, out of the timing constraints and computational capabilities. Fault Injections (FIs) are one of the most used approaches for evaluating the dependability of safety-critical systems. With billion-transistor hardware devices running trillion-parameter deep neural networks, injecting the entire fault universe is unfeasible. A widespread solution consists in performing statistical fault injections (SFIs), injecting a subset of faults to estimate a characteristic with an error margin and a confidence level. This research work presents an iterative SFI approach to estimate failure rates in convolutional neural networks (CNNs), i.e., the percentage of wrong predictions caused by random hardware faults affecting synaptic weights. SFIs at different granularities have been performed with margin of errors equal to 1%, 0.1%, and 0.01%. Results for two CNNs (ResNet20 and MobileNetV2) are presented and experimentally and statistically demonstrate the effectiveness of the proposed approach. For instance, to estimate the network-wise failure rate with an error margin of 0.01%, the proposed approach reduces the total injected faults by about 66% and 90% compared to conservative methods, and by 1.94% and 1.65% compared to iterative SFI methods in the literature, for ResNet20 and MobileNetV2, respectively.
10 February 2023 in Computer
To improve the trustworthiness of artificial intelligence systems, organizations should address the risks relevant to the use case whilst taking into account the stakeholders that interact directly or indirectly with the system.
15 March 2021 in Computer
3 April 2006 in SAE International Journal of Advances and Current Practices in Mobility
This paper presents a platform-based approach for the design of System-on-Chips for safety-critical microcontrollers, in which a set of HW and SW IPs, tools and methodologies are used for detection and correction of faults affecting the different parts of the electronic equipment. The IEC 61508 norm has been used as guideline and a systematic validation methodology with fault injection is presented. The paper shows an application of this approach to a PHILIPS reference platform, and first results are given in terms of benefits and costs.