GRADUATE SCHOOL

Ph.D. In Electrical-Electronics Engineering

IES 550 | Course Introduction and Application Information

Course Name
Artificial Neural Networks
Code
Semester
Theory
(hour/week)
Application/Lab
(hour/week)
Local Credits
ECTS
IES 550
Fall/Spring
3
0
3
7.5

Prerequisites
None
Course Language
English
Course Type
Elective
Course Level
Second Cycle
Mode of Delivery -
Teaching Methods and Techniques of the Course -
Course Coordinator -
Course Lecturer(s) -
Assistant(s) -
Course Objectives This course will introduce the fundamental principles and algorithms of Artificial Neural Network (ANN) systems. The course will cover many subjects including basic neuron model, simple perceptron, multilayer perceptron, Backpropagation learning algorithms; radialbasis function (RBF) networks; selforganizing maps (SOM) and learning vector quantization (LVQ); supportvector machines; classification techniques; pattern recognition.
Learning Outcomes The students who succeeded in this course;
  • Be able to describe basic artificial neural network models and how they relate to artificial intelligence
  • Have an understanding of the most common ANN architectures and their learning algorithms
  • Understand supervised vs. unsupervised learning techniques
  • Be able to discuss the main factors involved in achieving good learning and generalization performance in neural network systems
  • Be able to evaluate the practical considerations in applying ANNs to real classification and pattern recognition problems, basic ANNs algorithms using Matlab and its Neural Network Toolbox
Course Description The following topics will be included: The main neural network architectures and learning algorithms; Perceptrons and the LMS algorithm; Backpropagation learning; Recurrent networks; Radial basis functions; Pattern classification; Support vector machines; Kohonen’s selforganizing feature maps; Hopfield networks.

 



Course Category

Core Courses
Major Area Courses
Supportive Courses
Media and Management Skills Courses
Transferable Skill Courses

 

WEEKLY SUBJECTS AND RELATED PREPARATION STUDIES

Week Subjects Related Preparation
1 Yapay sinir ağlarına giriş, motivasyon ve uygulamaları /Introduction to artificial neural nets, motivation and applications Chapter 1. Sections 1.1. Neural Networks: Algorithms, Applications, and Programming Techniques. Freeman & Skapura. ISBN 0201513765
2 Temel sinir hücre modeli, Hebbian öğrenme kuralı, sinir ağları yapıları/Basic neuron model, Hebbian learning rule, neural network structures Chapter 1. Sections 1.31.6. Neural Networks: A Comprehensive Foundation. Haykin. ISBN 8178083000
3 Tekkatlı perceptron, doğrusal olmayan aktivasyon fonksiyonu, Delta kuralı /Singlelayer perceptron, nonlinear activation function, Delta rule Chapter 1. Sections 1.2. Neural Networks: Algorithms, Applications, and Programming Techniques. Freeman & Skapura. ISBN 0201513765
4 En küçükortalamalıkare (LMS) algoritması, perceptron yakınsama teoremi /The leastmeansquare (LMS) algorithm, perceptron convergence theorem Chapter 2. Sections 2.2. Neural Networks: Algorithms, Applications, and Programming Techniques. Freeman & Skapura. ISBN 0201513765Chapter 3. Sections 3.5, 3.9. Neural Networks: A Comprehensive Foundation. Haykin. ISBN 8178083000
5 Denetimli öğrenim: çokkatlı ağlar. Backpropagation Öğrenimi /Supervised learning: multilayer networks. Backpropagation Learning Chapter 3. Sections 3.1, 3.2. Neural Networks: Algorithms, Applications, and Programming Techniques. Freeman & Skapura. ISBN 0201513765Chapter 4. Sections 4.34.5. Neural Networks: A Comprehensive Foundation. Haykin. ISBN 8178083000
6 Ağ eğitimi ve testi için genel pratikler, çokkatlı perceptron uygulamaları /General practices for network training and testing, applications of multilayer perceptrons Chapter 3. Sections 3.3, 3.4. Neural Networks: Algorithms, Applications, and Programming Techniques. Freeman & Skapura. ISBN 0201513765
7 Özyineli Backprop Ağları; Zaman içinde backpropagation öğrenim algoritması /Recurrent Backprop networks; Backpropagation through time learning algorithm P.J. Werbos, “Backpropagation through time: what it does and how to do it,” Proceedings of the IEEE, 78(10), 15501560, 1990.
8 Radyaltabanlı fonksiyon (RBF) ağları ve düzenlileştirme teorisi /Radialbasis function (RBF) networks and regularization theory Chapter 5. Sections 5.7,5.8,5.10,5.11. Neural Networks: A Comprehensive Foundation. Haykin. ISBN 8178083000
9 Çekirdek metodları, destek vektör makinesi /Kernel methods, support vector machine Chapter 6. Sections 6.16.5. Neural Networks: A Comprehensive Foundation. Haykin. ISBN 8178083000
10 Uzmanların karışımı, EM (beklentienbüyütme) algoritması /Mixture of experts, the EM (ExpectationMaximization) algorithm Chapter 7. Sections 7.7,7.10,7.12. Neural Networks: A Comprehensive Foundation. Haykin. ISBN 8178083000
11 Denetimsiz öğrenim, ana bileşenler analizi, rekabetçi ağlar/ Unsupervised learning, principal components analysis, competitive networks Chapter 8. Sections 8.3. Neural Networks: A Comprehensive Foundation. Haykin. ISBN 8178083000
12 Hopfield ağları ve Boltzmann makineleri /Hopfield networks and Boltzmann machines Chapter 4 & 5. Sections 4.14.3, 5.1,5.2. Neural Networks: Algorithms, Applications, and Programming Techniques. Freeman & Skapura. ISBN 0201513765
13 Kohonen’nin kendi kendini örgütleyen haritaları: Algoritmalar ve uygulamalar /Kohonen’s selforganizing feature maps: Algorithms and applications Chapter 7. Sections 7.1,7.2. Neural Networks: Algorithms, Applications, and Programming Techniques. Freeman & Skapura. ISBN 0201513765
14 Bilgiteorik modeller, bağımsız bileşenler analizi /Informationtheoric models, independent components analysis Chapter 10. Sections 10.110.3, 10.11. Neural Networks: A Comprehensive Foundation. Haykin. ISBN 8178083000
15 Örüntü tanıma için sinir ağları / Neural networks for pattern recognition Chapter 1. Neural Networks for Pattern Recognition. Bishop. ISBN13 9780198538646
16 Review of the Semester  

 

Course Notes/Textbooks James A. Freeman and David M. Skapura, “Neural Networks: Algorithms, Applications, and Programming Techniques”, AddisonWesley Publishing Co., 1991, ISBN 0201513765.S. Haykin, “Neural Networks: A Comprehensive Foundation”, PrenticeHall, 2nd Ed., 1999, ISBN 8178083000.
Suggested Readings/Materials C. M. Bishop, “Neural Networks for Pattern Recognition”, Oxford University Press, 1996, ISBN13: 9780198538646.

 

EVALUATION SYSTEM

Semester Activities Number Weigthing
Participation
Laboratory / Application
Field Work
Quizzes / Studio Critiques
Portfolio
Homework / Assignments
Presentation / Jury
1*
30
Project
Seminar / Workshop
Oral Exams
Midterm
1
20
Final Exam
1
30
Total

Weighting of Semester Activities on the Final Grade
70
Weighting of End-of-Semester Activities on the Final Grade
30
Total

ECTS / WORKLOAD TABLE

Semester Activities Number Duration (Hours) Workload
Theoretical Course Hours
(Including exam week: 16 x total hours)
16
3
48
Laboratory / Application Hours
(Including exam week: '.16.' x total hours)
16
0
Study Hours Out of Class
15
6
90
Field Work
0
Quizzes / Studio Critiques
0
Portfolio
0
Homework / Assignments
0
Presentation / Jury
1*
5
5
Project
0
Seminar / Workshop
0
Oral Exam
0
Midterms
1
11
11
Final Exam
1
16
16
    Total
170

 

COURSE LEARNING OUTCOMES AND PROGRAM QUALIFICATIONS RELATIONSHIP

#
Program Competencies/Outcomes
* Contribution Level
1
2
3
4
5
1 Accesses information in breadth and depth by conducting scientific research in Electrical and Electronics Engineering; evaluates, interprets and applies information.
2 Is well-informed about contemporary techniques and methods used in Electrical and Electronics Engineering and their limitations.
3 Uses scientific methods to complete and apply information from uncertain, limited or incomplete data; can combine and use information from different disciplines. Knows and applies the research methods in studies of the area with a high level of skill.
4 Is informed about new and upcoming applications in the field and learns them whenever necessary.
5 Defines and formulates problems related to Electrical and Electronics Engineering, develops methods to solve them and uses progressive methods in solutions. Can independently realize novel studies that bring innovation to the field, or methods, or design, or known methods.
6 Develops novel and/or original methods, designs complex systems or processes and develops progressive/alternative solutions in designs.
7 Designs and implements studies based on theory, experiments and modeling; analyses and resolves the complex problems that arise in this process. Performs critical analysis, synthesis and evaluation of new and complex ideas.
8 Can work effectively in interdisciplinary teams as well as teams of the same discipline, can lead such teams and can develop approaches for resolving complex situations; can work independently and takes responsibility.
9 Engages in written and oral communication at least in Level C1 of the European Language Portfolio Global Scale.
10 Communicates the process and the results of his/her studies in national and international venues systematically, clearly and in written or oral form.
11 Evaluates the results of scientific, technological and engineering research and development activities in terms of the social, environmental, health, safety and legal aspects. Examines social relations and norms related to the field, and develops and makes attempts to change them if necessary. Knows their project management and business applications, and is aware of their limitations in Electrical and Electronics Engineering applications.
12 Highly regards scientific and ethical values in data collection, interpretation, communication and in every professional activity. Adheres to the principles of research and publication ethics.

*1 Lowest, 2 Low, 3 Average, 4 High, 5 Highest

 


Izmir University of Economics
is an establishment of
izto logo
Izmir Chamber of Commerce Health and Education Foundation.
ieu logo

Sakarya Street No:156
35330 Balçova - İzmir / Turkey

kampus izmir

Follow Us

İEU © All rights reserved.