An Energy Efficient Time-mode Digit Classification Neural Network Implementation

Abstract: This paper presents the design of an ultra-low energy neural network that utilizes time-mode signal processing (TMSP). Handwritten digit classification using a single-layer artificial neural network (ANN) with a Softmin-based activation function is described as an implementation example. To realize time mode operation, the presented design makes use of monostable multivibrator based multiplying analog- to-time converters, fixed-width pulse generators and basic digital gates. The time-mode digit classification ANN was designed in a standard CMOS 0.18 µm IC process and operates from a supply voltage of 0.6V. The system operates on the MNIST database of handwritten digits with quantized neuron weights and has a classification accuracy of 88%, which is typical for single-layer ANNs, while dissipating 65.74 pJ per classification with a speed of 2.37k.

Published in: Philosophical Transactions of the Royal Society A

DOI: https://doi.org/10.1098/rsta.2019.0163

Download PDF

@article{akgun2020energy,
 title={An Energy Efficient Time-mode Digit Classification Neural Network Implementation},
 author={Akgun, OC and Mei, J},
 journal={Philosophical Transactions of the Royal Society A},
 volume={378},
 number={2164},
 pages={20190163},
 year={2020},
 publisher={The Royal Society Publishing}
}