Jan, 2024
基于蒙特卡洛失活的贝叶斯神经网络施行旋转电子学测试
Testing Spintronics Implemented Monte Carlo Dropout-Based Bayesian Neural Networks
Soyed Tuhin Ahmed, Michael Hefenbrock, Guillaume Prenat, Lorena Anghel, Mehdi B. Tahoori
TL;DRBayesian Neural Networks (BayNNs) and uncertainty estimation in spintronics-based computation-in-memory architectures are analyzed with a focus on the reliability of Dropout generation and BayNN computation, proposing a testing framework for Dropout-based BayNN with high fault coverage and minimal training data usage.