Bringing uncertainty quantification to the extreme-edge with memristor-based Bayesian neural networks

D Bonnet, T Hirtzlin, A Majumdar, T Dalgaty… - Nature …, 2023 - nature.com
Safety-critical sensory applications, like medical diagnosis, demand accurate decisions from
limited, noisy data. Bayesian neural networks excel at such tasks, offering predictive …

A memristor-based Bayesian machine

KE Harabi, T Hirtzlin, C Turck, E Vianello, R Laurent… - Nature …, 2023 - nature.com
Memristors, and other emerging memory technologies, can be used to create energy-
efficient implementations of neural networks. However, for certain edge applications (in …

Improving realistic worst-case performance of NVCiM DNN accelerators through training with right-censored gaussian noise

Z Yan, Y Qin, W Wen, XS Hu… - 2023 IEEE/ACM …, 2023 - ieeexplore.ieee.org
Compute-in-Memory (CiM), built upon non-volatile memory (NVM) devices, is promising for
accelerating deep neural networks (DNNs) owing to its in-situ data processing capability …

Computing-in-memory neural network accelerators for safety-critical systems: Can small device variations be disastrous?

Z Yan, XS Hu, Y Shi - Proceedings of the 41st IEEE/ACM International …, 2022 - dl.acm.org
Computing-in-Memory (CiM) architectures based on emerging nonvolatile memory (NVM)
devices have demonstrated great potential for deep neural network (DNN) acceleration …

Reliable memristor-based neuromorphic design using variation-and defect-aware training

D Gaol, GL Zhang, X Yin, B Li… - 2021 IEEE/ACM …, 2021 - ieeexplore.ieee.org
The memristor crossbar provides a unique opportunity to develop a neuromorphic
computing system (NCS) with high scalability and energy efficiency. However, the reliability …

Enhancing reliability of neural networks at the edge: Inverted normalization with stochastic affine transformations

ST Ahmed, K Danouchi, G Prenat… - … , Automation & Test …, 2024 - ieeexplore.ieee.org
Bayesian Neural Networks (BayNNs) naturally provide uncertainty in their predictions,
making them a suitable choice in safety-critical applications. Additionally, their realization …

Brocom: A bayesian framework for robust computing on memristor crossbar

D Gao, Z Yang, Q Huang, GL Zhang… - … on Computer-Aided …, 2022 - ieeexplore.ieee.org
Memristor crossbar arrays are considered to be a promising platform for neuromorphic
computing. To deploy a trained neural network (NN) model on memristor crossbars …

AD2VNCS: Adversarial Defense and Device Variation-tolerance in Memristive Crossbar-based Neuromorphic Computing Systems

Y Bi, Q Xu, H Geng, S Chen, Y Kang - ACM Transactions on Design …, 2023 - dl.acm.org
In recent years, memristive crossbar-based neuromorphic computing systems (NCS) have
obtained extremely high performance in neural network acceleration. However, adversarial …

On the reliability of computing-in-memory accelerators for deep neural networks

Z Yan, XS Hu, Y Shi - … System Dependability from Data, System and …, 2022 - Springer
Computing-in-memory with emerging non-volatile memory (nvCiM) is shown to be a
promising candidate for accelerating deep neural networks (DNNs) with high energy …

Negative Feedback Training: A Novel Concept to Improve Robustness of NVCiM DNN Accelerators

Y Qin, Z Yan, W Wen, XS Hu, Y Shi - arXiv preprint arXiv:2305.14561, 2023 - arxiv.org
Compute-in-memory (CIM) accelerators built upon non-volatile memory (NVM) devices
excel in energy efficiency and latency when performing Deep Neural Network (DNN) …