CIVILICA We Respect the Science
(ناشر تخصصی کنفرانسهای کشور / شماره مجوز انتشارات از وزارت فرهنگ و ارشاد اسلامی: ۸۹۷۱)

Deep learning, concepts and performance: survey

عنوان مقاله: Deep learning, concepts and performance: survey
شناسه ملی مقاله: ICELE05_123
منتشر شده در پنجمین کنفرانس ملی مهندسی برق و مکاترونیک ایران در سال 1398
مشخصات نویسندگان مقاله:

Mahdieh Abdipoor - Islamic Azad University of Najaf Abad, Iran
Hamid Salehi - Islamic Azad University of Shiraz, Iran

خلاصه مقاله:
Deep learning is a branch of machine learning concept and a set of algorithms that attempts to model high-level abstract concepts using learning at different levels and layers. In deep learning, artificial neural networks and algorithms similar to the human brain learn skills from massive datasets. The reason for using deep learning is that neural networks have different or deep layers that make learning possible. The purpose of this paper is to thoroughly review deep learning and its concepts, applications and classification, as well as provide a brief description of how to create and teach deep learning models and implement it in MATLAB software. We hope this article is a valuable reference for new researchers of deep learning and also people who seek for innovation in deep learning applications.

کلمات کلیدی:
deep learning, neural networks, machine learning, deep learning with MATLAB.

صفحه اختصاصی مقاله و دریافت فایل کامل: https://civilica.com/doc/988454/