Utilize the web fullyThanks to cookies, browsing the web will be easier and more comfortable. To process some of them, we need your consent which you can give by clicking the “Accept all” button. More information about cookies and their settings. By refusing, only essential cookies will be loaded.

Accept all Refuse

UO > Research > Projects & Grants

Projects & Grants

Data and image analysis
Project Id	SGS01/ÚVAFM/20
Main solver	MgA. Jan Hůla, Ph.D.
Period	1/2020 - 12/2020
Provider	Specifický VŠ výzkum
State	finished
Anotation	This project thematically follows previous SGS titled "Preprocessing methods and techniques in data analysis" and at the same time is extended by particular directions related to Neural Networks. We will continue research in Dimensionality Reduction, various notions of similarity and Image processing. In addition, we will focus on a new type of neural networks designed to work with structured inputs, analysis of the influence of error function on model behavior and optimization of the architecture of pre-trained neural networks. Our theoretical and practical results will be presented at international and local conferences.