التفاصيل البيبلوغرافية
العنوان: |
1S1R Optimization for High‐Frequency Inference on Binarized Spiking Neural Networks. |
المؤلفون: |
Minguet Lopez, Joel, Rafhay, Quentin, Dampfhoffer, Manon, Reganaz, Lucas, Castellani, Niccolo, Meli, Valentina, Martin, Simon, Grenouillet, Laurent, Navarro, Gabriele, Magis, Thomas, Carabasse, Catherine, Hirtzlin, Tifenn, Vianello, Elisa, Deleruyelle, Damien, Portal, Jean‐Michel, Molas, Gabriel, Andrieu, François |
المصدر: |
Advanced Electronic Materials; Aug2022, Vol. 8 Issue 8, p1-11, 11p |
مصطلحات موضوعية: |
ARTIFICIAL neural networks, MONTE Carlo method, ARTIFICIAL intelligence, DEEP learning, NONVOLATILE memory, IMAGE recognition (Computer vision), MACHINE learning |
مستخلص: |
Single memristor crossbar arrays are a very promising approach to reduce the power consumption of deep learning accelerators. In parallel, the emerging bio‐inspired spiking neural networks (SNNs) offer very low power consumption with satisfactory performance on complex artificial intelligence tasks. In such neural networks, synaptic weights can be stored in nonvolatile memories. The latter are massively read during inference, which can lead to device failure. In this context, a 1S1R (1 Selector 1 Resistor) device composed of a HfO2‐based OxRAM memory stacked on a Ge‐Se‐Sb‐N‐based ovonic threshold switch (OTS) back‐end selector is proposed for high‐density binarized SNNs (BSNNs) synaptic weight hardware implementation. An extensive experimental statistical study combined with a novel Monte Carlo model allows to deeply analyze the OTS switching dynamics based on field‐driven stochastic nucleation of conductive dots in the layer. This allows quantifying the occurrence frequency of OTS erratic switching as a function of the applied voltages and 1S1R reading frequency. The associated 1S1R reading error rate is calculated. Focusing on the standard machine learning MNIST image recognition task, BSNN figures of merit (footprint, electrical consumption during inference, frequency of inference, accuracy, and tolerance to errors) are optimized by engineering the network topology, training procedure, and activations sparsity. [ABSTRACT FROM AUTHOR] |
|
Copyright of Advanced Electronic Materials is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) |
قاعدة البيانات: |
Complementary Index |