Improving Neural Network Efficiency Using Piecewise Linear Approximation of Activation Functions

Activation functions play a pivotal role in Neural Networks by enabling the modeling of complex non-linear relationships within data. However, the computational cost associated with certain activation functions, such as the hyperbolic tangent (tanh) and its gradient, can be substantial. In this stu...

Full description

Saved in:
Bibliographic Details
Main Authors: Pavan Reddy, Aditya Sanjay Gujral
Format: Article
Language:English
Published: LibraryPress@UF 2025-05-01
Series:Proceedings of the International Florida Artificial Intelligence Research Society Conference
Online Access:https://journals.flvc.org/FLAIRS/article/view/139005
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850135456777240576
author Pavan Reddy
Aditya Sanjay Gujral
author_facet Pavan Reddy
Aditya Sanjay Gujral
author_sort Pavan Reddy
collection DOAJ
description Activation functions play a pivotal role in Neural Networks by enabling the modeling of complex non-linear relationships within data. However, the computational cost associated with certain activation functions, such as the hyperbolic tangent (tanh) and its gradient, can be substantial. In this study, we demonstrate that a piecewise linear approximation of the tanh function, utilizing pre-calculated slopes, achieves faster computation without significant degradation in performance. Conversely, we show that a piecewise linear approximation of the sigmoid function is computationally slower compared to its continuous counterpart. These findings suggest that the computational efficiency of a piecewise activation function depends on whether the indexing and arithmetic costs of the approximation are lower than those of the continuous function.
format Article
id doaj-art-84f6ae1f5cc948188f24034f53a4cd62
institution OA Journals
issn 2334-0754
2334-0762
language English
publishDate 2025-05-01
publisher LibraryPress@UF
record_format Article
series Proceedings of the International Florida Artificial Intelligence Research Society Conference
spelling doaj-art-84f6ae1f5cc948188f24034f53a4cd622025-08-20T02:31:26ZengLibraryPress@UFProceedings of the International Florida Artificial Intelligence Research Society Conference2334-07542334-07622025-05-0138110.32473/flairs.38.1.139005Improving Neural Network Efficiency Using Piecewise Linear Approximation of Activation FunctionsPavan Reddy0https://orcid.org/0009-0001-4832-1845Aditya Sanjay Gujral1The George Washington UniversityThe George Washington University Activation functions play a pivotal role in Neural Networks by enabling the modeling of complex non-linear relationships within data. However, the computational cost associated with certain activation functions, such as the hyperbolic tangent (tanh) and its gradient, can be substantial. In this study, we demonstrate that a piecewise linear approximation of the tanh function, utilizing pre-calculated slopes, achieves faster computation without significant degradation in performance. Conversely, we show that a piecewise linear approximation of the sigmoid function is computationally slower compared to its continuous counterpart. These findings suggest that the computational efficiency of a piecewise activation function depends on whether the indexing and arithmetic costs of the approximation are lower than those of the continuous function. https://journals.flvc.org/FLAIRS/article/view/139005
spellingShingle Pavan Reddy
Aditya Sanjay Gujral
Improving Neural Network Efficiency Using Piecewise Linear Approximation of Activation Functions
Proceedings of the International Florida Artificial Intelligence Research Society Conference
title Improving Neural Network Efficiency Using Piecewise Linear Approximation of Activation Functions
title_full Improving Neural Network Efficiency Using Piecewise Linear Approximation of Activation Functions
title_fullStr Improving Neural Network Efficiency Using Piecewise Linear Approximation of Activation Functions
title_full_unstemmed Improving Neural Network Efficiency Using Piecewise Linear Approximation of Activation Functions
title_short Improving Neural Network Efficiency Using Piecewise Linear Approximation of Activation Functions
title_sort improving neural network efficiency using piecewise linear approximation of activation functions
url https://journals.flvc.org/FLAIRS/article/view/139005
work_keys_str_mv AT pavanreddy improvingneuralnetworkefficiencyusingpiecewiselinearapproximationofactivationfunctions
AT adityasanjaygujral improvingneuralnetworkefficiencyusingpiecewiselinearapproximationofactivationfunctions