Kademeli evrişimli sinir ağlarında uyarlanabilir ağ seçimi tekniği

Loading...
Thumbnail Image

Date

2023

Journal Title

Journal ISSN

Volume Title

Publisher

Open Access Color

OpenAIRE Downloads

OpenAIRE Views

Research Projects

Organizational Units

Organizational Unit
Software Engineering
(2005)
Department of Software Engineering was founded in 2005 as the first department in Ankara in Software Engineering. The recent developments in current technologies such as Artificial Intelligence, Machine Learning, Big Data, and Blockchains, have placed Software Engineering among the top professions of today, and the future. The academic and research activities in the department are pursued with qualified faculty at Undergraduate, Graduate and Doctorate Degree levels. Our University is one of the two universities offering a Doctorate-level program in this field. In addition to focusing on the basic phases of software (analysis, design, development, testing) and relevant methodologies in detail, our department offers education in various areas of expertise, such as Object-oriented Analysis and Design, Human-Computer Interaction, Software Quality Assurance, Software Requirement Engineering, Software Design and Architecture, Software Project Management, Software Testing and Model-Driven Software Development. The curriculum of our Department is catered to graduate individuals who are prepared to take part in any phase of software development of large-scale software in line with the requirements of the software sector. Department of Software Engineering is accredited by MÜDEK (Association for Evaluation and Accreditation of Engineering Programs) until September 30th, 2021, and has been granted the EUR-ACE label that is valid in Europe. This label provides our graduates with a vital head-start to be admitted to graduate-level programs, and into working environments in European Union countries. The Big Data and Cloud Computing Laboratory, as well as MobiLab where mobile applications are developed, SimLAB, the simulation laboratory for Medical Computing, and software education laboratories of the department are equipped with various software tools and hardware to enable our students to use state-of-the-art software technologies. Our graduates are employed in software and R&D companies (Technoparks), national/international institutions developing or utilizing software technologies (such as banks, healthcare institutions, the Information Technologies departments of private and public institutions, telecommunication companies, TÜİK, SPK, BDDK, EPDK, RK, or universities), and research institutions such TÜBİTAK.

Journal Issue

Abstract

Dinamik sinir ağı, derin öğrenmede önemli bir araştırma alanıdır. Sunulan tez, statik modellerin verimliliğini ve uyarlanabilirliğini artırmak için iki veya daha fazla sinir ağını artan derinlikte bağlamak için bir yönlendirici kullanan kademeli sinir ağına odaklanmaktadır. Bu tezde, kademeli derin sinir ağlarında ağ seçimi için parametresiz bir teknik önerdik. Bu teknik, sığ ağların da birçok örneği doğru bir şekilde sınıflandırabilmesi gerçeğinden yararlanarak, eğitim ve çıkarım için gereken hesaplama süresini azaltmayı amaçlamaktadır. Kademeli sinir ağı, softmax marjı ve klasik LeNet modelinin kısa bir açıklamasını takiben, yeni bir kademeli sinir ağı algoritması tanıtılmaktadır. Önerilen model; MNIST, EMNIST ve Fashion-MNIST veri kümelerinde etkinlik ve performans açısından LeNet ile karşılaştırılmaktadır. Sayısal sonuçlar, önerilen teknikle referans modelinin verimliliğinin büyük ölçüde arttığını ve doğruluktan ödün vermeden geliştirildiğini göstermektedir.
Dynamic neural network is an important research area in deep learning. The presented thesis focuses on cascaded neural network which is a sub-topic of dynamic neural network, that utilizes a router for connecting two or more neural networks with increasing depth in order to enhance the efficiency and adaptiveness of static models. In this thesis, we proposed a parameter-free technique for network selection in cascaded deep neural networks in order to reduce the computational time required for training and inference by taking advantage of the fact that shallow networks are also able to correctly classify many samples. Following a brief explanation of the cascaded neural network, softmax margin, and classical LeNet model; a novel cascaded neural network algorithm is introduced. The proposed model is compared to LeNet in terms of efficiency and performance on MNIST, EMNIST, and Fashion-MNIST datasets with various networks utilized as small modules of the cascaded model. Numerical results demonstrated that the proposed technique greatly improves the efficiency of the benchmark model without sacrificing accuracy.

Description

Keywords

Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol, Computer Engineering and Computer Science and Control

Turkish CoHE Thesis Center URL

Fields of Science

Citation

WoS Q

Scopus Q

Source

Volume

Issue

Start Page

0

End Page

67