Testing Spintronics Implemented Monte Carlo Dropout-Based Bayesian Neural Networks
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929205800861696 |
|---|---|
| author | Ahmed, Soyed Tuhin Hefenbrock, Michael Prenat, Guillaume Anghel, Lorena Tahoori, Mehdi B. |
| author_facet | Ahmed, Soyed Tuhin Hefenbrock, Michael Prenat, Guillaume Anghel, Lorena Tahoori, Mehdi B. |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_04744 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Testing Spintronics Implemented Monte Carlo Dropout-Based Bayesian Neural Networks Ahmed, Soyed Tuhin Hefenbrock, Michael Prenat, Guillaume Anghel, Lorena Tahoori, Mehdi B. Emerging Technologies Hardware Architecture Machine Learning 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. |
| title | Testing Spintronics Implemented Monte Carlo Dropout-Based Bayesian Neural Networks |
| topic | Emerging Technologies Hardware Architecture Machine Learning |
| url | https://arxiv.org/abs/2401.04744 |