Genetic Algorithm for Automatic tuning of neural network hyperparameters

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F63839172%3A_____%2F18%3A10133127" target="_blank" >RIV/63839172:_____/18:10133127 - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.1117/12.2304955" target="_blank" >http://dx.doi.org/10.1117/12.2304955</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1117/12.2304955" target="_blank" >10.1117/12.2304955</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Genetic Algorithm for Automatic tuning of neural network hyperparameters

Popis výsledku v původním jazyce

Articial neural networks affect our everyday life. But every neural network depends on the appropriate training set and setting of internal properties with hyperparameters. Even accurate and complete training set doesnt imply high performance of neural network algorithm. Tuning of hyperparameters is crucial for correct functionality, fast learning and high accuracy of neural networks. The hyperparameter selection relies on manual ne-tuning based on multiple full training trials. There are a lot of neural network implementation available for public and commercial use, but the setting of hyperparameters is often a neglected problem. Choosing the best structure and hyperparameters is the primary challenge in designing a neural network. This article describes a genetic algorithm for automatic selection of hyperparameters and their tuning for increasing the performance of neural networks without human interaction. The optimization algorithm accelerates the discovery of conguration, which is otherwise a time-consuming task. We evaluate the results of optimizations in comparison to naive approach and compare pro and cons of different techniques.

Název v anglickém jazyce

Genetic Algorithm for Automatic tuning of neural network hyperparameters

Popis výsledku anglicky

Articial neural networks affect our everyday life. But every neural network depends on the appropriate training set and setting of internal properties with hyperparameters. Even accurate and complete training set doesnt imply high performance of neural network algorithm. Tuning of hyperparameters is crucial for correct functionality, fast learning and high accuracy of neural networks. The hyperparameter selection relies on manual ne-tuning based on multiple full training trials. There are a lot of neural network implementation available for public and commercial use, but the setting of hyperparameters is often a neglected problem. Choosing the best structure and hyperparameters is the primary challenge in designing a neural network. This article describes a genetic algorithm for automatic selection of hyperparameters and their tuning for increasing the performance of neural networks without human interaction. The optimization algorithm accelerates the discovery of conguration, which is otherwise a time-consuming task. We evaluate the results of optimizations in comparison to naive approach and compare pro and cons of different techniques.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/EF16_013%2F0001797" target="_blank" >EF16_013/0001797: E-infrastruktura CESNET - modernizace</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2018

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Autonomous Systems: Sensors, Vehicles, Security, and the Internet of Everything

ISBN

978-1-5106-1798-8

ISSN

0277-786X

e-ISSN

neuvedeno

Počet stran výsledku

7

Strana od-do

—

Název nakladatele

SPIE

Místo vydání

Neuveden

Místo konání akce

Orlando, Florida, United States

Datum konání akce

16. 4. 2018

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000453556800017