Convolutional neural network as an architecture for deep learning

Kocaleva, Mirjana (2018) Convolutional neural network as an architecture for deep learning. In: Научна конференция на младите изследователи, 18 May 2018, Veliko Trnovo, Bugarija.

[thumbnail of 39.Mladi_izsledovateli_2018.pdf]
Preview
Text
39.Mladi_izsledovateli_2018.pdf

Download (975kB) | Preview

Abstract

Machine learning has become important for solving problems in many areas: computational finance, image processing and computer vision, face recognition, motion detection, object detection, tumour detection, drug discovery, DNA sequencing, energy production, price and load forecasting, automotive, aerospace, manufacturing, predictive maintenance, etc. The paper reviews the machine learning as a process for teaching computers to learn from experience or directly from data without being based on a predefined equation as a model. Then neural networks are considered. Special attention is paid to the deep learning and some architecture for deep learning are reviewed. For this research and further research the most important architecture for deep learning is convolutional neural network and we will hold on to it.

Item Type: Conference or Workshop Item (Paper)
Subjects: Natural sciences > Computer and information sciences
Divisions: Faculty of Computer Science
Depositing User: Mirjana Kocaleva Vitanova
Date Deposited: 20 Jun 2019 07:19
Last Modified: 20 Jun 2019 07:19
URI: https://eprints.ugd.edu.mk/id/eprint/20165

Actions (login required)

View Item View Item