Please use this identifier to cite or link to this item:
https://dspace.uzhnu.edu.ua/jspui/handle/lib/74865Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.author | Коцовський, Владислав | - |
| dc.date.accessioned | 2025-06-29T17:55:39Z | - |
| dc.date.available | 2025-06-29T17:55:39Z | - |
| dc.date.issued | 2024 | - |
| dc.identifier.uri | https://dspace.uzhnu.edu.ua/jspui/handle/lib/74865 | - |
| dc.description.abstract | The modification of the online learning algorithm for multi-valued multithreshold neurons is proposed in the paper. Conditions are stated and proved that ensure the finite successful learning. The influence of the algorithm hyperparameters on the learning process is analyzed on the base of simulation results. The recommendations are formulated concerning the choice of values of these hyperparameters, which may significantly reduce the learning time. | uk |
| dc.language.iso | en | uk |
| dc.subject | Multithreshold neural unit, classification, machine learning | uk |
| dc.title | Hyperparameter tuning in the learning of multithreshold neurons | uk |
| dc.type | Presentation | uk |
| dc.pubType | Тези до статті | uk |
| Appears in Collections: | Наукові публікації кафедри інформаційних управляючих систем та технологій | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| presentation_68.pptx | 11.22 MB | Microsoft Powerpoint XML | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.