Testing Spintronics Implemented Monte Carlo Dropout-Based Bayesian Neural Networks
This addresses reliability issues in safety-critical applications using spintronics-based computation-in-memory architectures, representing an incremental advancement in testing methods.
The paper tackled the challenge of testing Bayesian Neural Networks implemented with spintronics-based dropout modules by modeling non-idealities and proposing a testing framework that achieves up to 100% fault coverage using only 0.2% of training data as test vectors.
Bayesian Neural Networks (BayNNs) can inherently estimate predictive uncertainty, facilitating informed decision-making. Dropout-based BayNNs are increasingly implemented in spintronics-based computation-in-memory architectures for resource-constrained yet high-performance safety-critical applications. Although uncertainty estimation is important, the reliability of Dropout generation and BayNN computation is equally important for target applications but is overlooked in existing works. However, testing BayNNs is significantly more challenging compared to conventional NNs, due to their stochastic nature. In this paper, we present for the first time the model of the non-idealities of the spintronics-based Dropout module and analyze their impact on uncertainty estimates and accuracy. Furthermore, we propose a testing framework based on repeatability ranking for Dropout-based BayNN with up to $100\%$ fault coverage while using only $0.2\%$ of training data as test vectors.