Catherine D. Schuman, James S. Plank, Grant Bruer and Jeremy Anantharaj

July, 2019

IJCNN: The International Joint Conference on Neural Networks

https://www.ijcnn.org/

View Article

Abstract

A key challenge for utilizing spiking neural networks or spiking neuromorphic systems for most applications is translating numerical data into spikes that are appropriate to apply as input to a spiking neural network. In this work, we present several approaches for encoding numerical values as spikes, including binning, spike-count encoding, and charge-injection encoding, and we show how these approaches can be combined hierarchically to form more complex encoding schemes. We demonstrate how these different encoding approaches perform on four different applications, running on four different neuromorphic systems that are based on spiking neural networks. We show that the input encoding method can have a significant effect on application performance and that the best input encoding method is application-specific.

Citation Information

Text

author      C. D. Schuman and J. S. Plank and G. Bruer and J. Anantharaj
title       Non-Traditional Input Encoding Schemes for Spiking Neuromorphic Systems
booktitle   IJCNN: The International Joint Conference on Neural Networks
year        2019
address     Budapest
pages       1-10
doi         10.1109/IJCNN.2019.8852139

Bibtex

@INPROCEEDINGS{spb:19:nte,
    author = "C. D. Schuman and J. S. Plank and G. Bruer and J. Anantharaj",
    title = "Non-Traditional Input Encoding Schemes for Spiking Neuromorphic Systems",
    booktitle = "IJCNN: The International Joint Conference on Neural Networks",
    year = "2019",
    address = "Budapest",
    pages = "1-10",
    doi = "10.1109/IJCNN.2019.8852139"
}