PAÜ .:. Eğitim Öğretim Bilgi Sistemi

COURSE INFORMATION

Course Code	Course Title	L+P Hour	Semester	ECTS
CENG 410	ARTIFICIAL NEURAL NETWORKS	3 + 0	8th Semester	5

COURSE DESCRIPTION

Course Level	Bachelor's Degree
Course Type	Elective
Course Objective	This course aims to give the basics of artificial neural network architectures and learning rules.
Course Content	Definition of artificial neural networks (ANN) ADALINE: adaptive linear element, Learning: supervised and unsupervised learning Linear Associative Memory, Multi-layer perceptron Back-propagation method, Radial-basis ANN Dynamic ANN, Hopfield Network Cellular ANN.
Prerequisites	No the prerequisite of lesson.
Corequisite	No the corequisite of lesson.
Mode of Delivery	Face to Face

COURSE LEARNING OUTCOMES

1	Learning articial neural networks
2	Understanding the application areas
3	Enhancing the subject area by a sample project on an application study

COURSE'S CONTRIBUTION TO PROGRAM

	PO 01	PO 02	PO 03	PO 04	PO 05	PO 06	PO 07	PO 08	PO 09	PO 10	PO 11	PO 12
LO 001	5	4	4	5	5		4			5	4
LO 002	5	4	4	5	5		4			5	4
LO 003	4	5	4	5	5		4			5	4
Sub Total	14	13	12	15	15		12			15	12
Contribution	5	4	4	5	5	0	4	0	0	5	4	0

ECTS ALLOCATED BASED ON STUDENT WORKLOAD BY THE COURSE DESCRIPTION

Activities	Quantity	Duration (Hour)	Total Work Load (Hour)
Course Duration (14 weeks/theoric+practical)	14	2	28
Hours for off-the-classroom study (Pre-study, practice)	14	2	28
Assignments	5	4	20
Mid-terms	1	24	24
Final examination	1	30	30
Total Work Load ECTS Credit of the Course			130 5

COURSE DETAILS

Select Year

Course Details

Course Code	Course Title	L+P Hour	Course Code	Language Of Instruction	Course Semester
CENG 410	ARTIFICIAL NEURAL NETWORKS	3 + 0	1	Turkish	2021-2022 Spring

Course Coordinator	E-Mail	Phone Number	Course Location	Attendance
Assoc. Prof. Dr. MERİÇ ÇETİN	mcetin@pau.edu.tr		MUH A0234	%70

Goals		This course aims to give the basics of artificial neural network architectures and learning rules.
Content		Definition of artificial neural networks (ANN) ADALINE: adaptive linear element, Learning: supervised and unsupervised learning Linear Associative Memory, Multi-layer perceptron Back-propagation method, Radial-basis ANN Dynamic ANN, Hopfield Network Cellular ANN.

Topics

Weeks	Topics
1	Introduction, Neuron Model and Network Architectures
2	Perceptron Learning Rule
3	Signal and Weight Vector Spaces, Linear Transformations for Neural Networks
4	Supervised Hebbian Learning
5	Performance Optimization
6	Widrow-Hoff Learning
7	Backpropagation
8	Variations on Backpropagation (Drawbacks of Backpropagation, Heuristic Modifications of Backpropagation, Momentum,Conjugate Gradient,Levenberg-Marquardt Algorithm)
9	Midterm
10	Generalization
11	Dynamic Networks
12	Associative Learning (Kohonen Rule)
13	Radial Basis Networks
14	Project Demonstrations

Materials

Materials are not specified.

Resources

Resources	Resources Language
Neural Network Design, Martin T. Hagan	Türkçe

Course Assessment

Assesment Methods	Percentage (%)	Assesment Methods Title
Final Exam	60	Final Exam
Midterm Exam	40	Midterm Exam

L+P: Lecture and Practice
PQ: Program Learning Outcomes
LO: Course Learning Outcomes

Prospective Student