English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In this paper, we provide a systematic approach for assessing and comparingthe computational complexity of neural network layers in digital signalprocessing. We provide and link four software-to-hardware complexity measures,defining how the different complexity metrics relate to the layers'hyper-parameters. This paper explains how to compute these four metrics forfeed-forward and recurrent layers, and defines in which case we ought to use aparticular metric depending on whether we characterize a more soft- orhardware-oriented application. One of the four metrics, called `the number ofadditions and bit shifts (NABS)', is newly introduced for heterogeneousquantization. NABS characterizes the impact of not only the bitwidth used inthe operation but also the type of quantization used in the arithmeticaloperations. We intend this work to serve as a baseline for the different levels(purposes) of complexity estimation related to the neural networks' applicationin real-time digital signal processing, aiming at unifying the computationalcomplexity estimation.

Computational Complexity Evaluation of Neural Network Applications in Signal Processing