Computationally efficient classification system architecture is proposed. It utilizes fast tensor-vector multiplication algorithm to apply linear operators upon input signals . The approach is applicable to wide variety of recognition system architectures ranging from single stage matched filter bank classifiers to complex neural networks with unlimited number of hidden layers.
@article{arxiv.1603.01772,
title = {Fast calculation of correlations in recognition systems},
author = {Pavel Dourbal and Mikhail Pekker},
journal= {arXiv preprint arXiv:1603.01772},
year = {2016}
}