Showing 1–1 of 1 results for author: Van Heukelom, M S

Using Floating Gate Memory to Train Ideal Accuracy Neural Networks

Authors: Sapan Agarwal, Diana Garland, John Niroula, Robin B, Jacobs-Gedrim, Alex Hsia, Michael S. Van Heukelom, Elliot Fuller, Bruce Draper, Matthew J. Marinella

Abstract: Floating gate SONOS (Silicon-Oxygen-Nitrogen-Oxygen-Silicon) transistors can be used to train neural networks to ideal accuracies that match those of floating point digital weights on the MNIST dataset when using multiple devices to represent a weight or within 1% of ideal accuracy when using a single device. This is enabled by operating devices in the subthreshold regime, where they exhibit symme… ▽ More Floating gate SONOS (Silicon-Oxygen-Nitrogen-Oxygen-Silicon) transistors can be used to train neural networks to ideal accuracies that match those of floating point digital weights on the MNIST dataset when using multiple devices to represent a weight or within 1% of ideal accuracy when using a single device. This is enabled by operating devices in the subthreshold regime, where they exhibit symmetric write nonlinearities. A neural training accelerator core based on SONOS with a single device per weight would increase energy efficiency by 120X, operate 2.1X faster and require 5X lower area than an optimized SRAM based ASIC. △ Less

Submitted 27 February, 2019; v1 submitted 29 January, 2019; originally announced January 2019.