Optimizing Correlation Algorithms for Hardware-Based Transient Classification
Abstract
The perfonnance of dedicated VLSI neural processing hardware depends critically on the design of the implemented algorithms. We have pre(cid:173) viously proposed an algorithm for acoustic transient classification [1]. Having implemented and demonstrated this algorithm in a mixed-mode architecture, we now investigate variants on the algorithm, using time and frequency channel differencing, input and output nonnalization, and schemes to binarize and train the template values, with the goal of achiev(cid:173) ing optimal classification perfonnance for the chosen hardware.
Cite
Text
Edwards et al. "Optimizing Correlation Algorithms for Hardware-Based Transient Classification." Neural Information Processing Systems, 1998.Markdown
[Edwards et al. "Optimizing Correlation Algorithms for Hardware-Based Transient Classification." Neural Information Processing Systems, 1998.](https://mlanthology.org/neurips/1998/edwards1998neurips-optimizing/)BibTeX
@inproceedings{edwards1998neurips-optimizing,
title = {{Optimizing Correlation Algorithms for Hardware-Based Transient Classification}},
author = {Edwards, R. Timothy and Cauwenberghs, Gert and Pineda, Fernando J.},
booktitle = {Neural Information Processing Systems},
year = {1998},
pages = {678-684},
url = {https://mlanthology.org/neurips/1998/edwards1998neurips-optimizing/}
}