Adapting the Size of Artificial Neural Networks Using Dynamic Auto-Sizing

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21240%2F22%3A00363052" target="_blank" >RIV/68407700:21240/22:00363052 - isvavai.cz</a>

Výsledek na webu

<a href="https://ieeexplore.ieee.org/abstract/document/10000471" target="_blank" >https://ieeexplore.ieee.org/abstract/document/10000471</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/CSIT56902.2022.10000471" target="_blank" >10.1109/CSIT56902.2022.10000471</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Adapting the Size of Artificial Neural Networks Using Dynamic Auto-Sizing

Popis výsledku v původním jazyce

We introduce dynamic auto-sizing, a novel approach to training artificial neural networks which allows the models to automatically adapt their size to the problem domain. The size of the models can be further controlled during the learning process by modifying the applied strength of regularization. The ability of dynamic auto-sizing models to expand or shrink their hidden layers is achieved by periodically growing and pruning entire units such as neurons or filters. For this purpose, we introduce weighted L1 regularization, a novel regularization method for inducing structured sparsity. Besides analyzing the behavior of dynamic auto-sizing, we evaluate predictive performance of models trained using the method and show that such models can provide a predictive advantage over traditional approaches.

Název v anglickém jazyce

Adapting the Size of Artificial Neural Networks Using Dynamic Auto-Sizing

Popis výsledku anglicky

We introduce dynamic auto-sizing, a novel approach to training artificial neural networks which allows the models to automatically adapt their size to the problem domain. The size of the models can be further controlled during the learning process by modifying the applied strength of regularization. The ability of dynamic auto-sizing models to expand or shrink their hidden layers is achieved by periodically growing and pruning entire units such as neurons or filters. For this purpose, we introduce weighted L1 regularization, a novel regularization method for inducing structured sparsity. Besides analyzing the behavior of dynamic auto-sizing, we evaluate predictive performance of models trained using the method and show that such models can provide a predictive advantage over traditional approaches.

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

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2022

Kód důvěrnosti údajů

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

Název statě ve sborníku

IEEE 17th International Conference on Computer Science and Information Technologies

ISBN

979-8-3503-3431-9

ISSN

—

e-ISSN

—

Počet stran výsledku

5

Strana od-do

592-596

Název nakladatele

IEEE

Místo vydání

Dortmund

Místo konání akce

Lvov

Datum konání akce

10. 11. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000927642900139