Course Code	Semester	Course Name	LE/RC/LA	Course Type	Language of Instruction	ECTS
YMB0015		Machine Learning and Neural Networks	3/0/0	DE	Turkish	7

Course Goals
Prerequisite(s)	None
Corequisite(s)	None
Special Requisite(s)	The minimum qualifications that are expected from the students who want to attend the course.(Examples: Foreign language level, attendance, known theoretical pre-qualifications, etc.)
Instructor(s)	Assist. Prof. Dr. Mehmet Fatih Uçar
Course Assistant(s)	None
Schedule	TBA
Office Hour(s)	Assist.Prof.Dr. Levent CUHACI, office phone : 4359
Teaching Methods and Techniques	-
Principle Sources	-
Other Sources	-

Course Schedules
Week	Contents	Learning Methods
1. Week	Machine Learning Principles	Lecture
2. Week	Target Functions and Inductive Bias	Lecture
3. Week	Decision Trees, ID3 and C4.5	Lecture
4. Week	Lazy Learning, CBR and K-Nearest Neighbours	Lecture
5. Week	Biological Neurons and Hebbian Learning	Lecture
6. Week	Perceptrons and Decision Spaces	Lecture
7. Week	Multi Layer-Perceptrons, and Back-Propagation Learning.	Lecture
8. Week	Advanced optimization algorithms for multilayer perceptron networks: conjugate gradient algorithm, Levenberg-Marquardt algorithm.	Lecture
9. Week	Recurrent Neural Networks	Lecture
10. Week	Midterm Exam	Exam
11. Week	Hopfield Neural Networks	Lecture
12. Week	Self-organizing Maps (SOM) and Learning Vector Quantization (LVQ)	Lecture
13. Week	Stochastic Networks	Lecture
14. Week	Evolutionary and Genetic Computation.	Lecture
15. Week	Final Exams Week	Exam
16. Week	Final Exams Week	Exam
17. Week	Final Exams Week	Exam

Assessments
Evaluation tools	Quantity	Weight(%)
Midterm(s)	1	30
Homework / Term Projects / Presentations	1	10
Attendance	1	10
Final Exam	1	50

Program Outcomes PO-1 Have scientific research in mathematics and computer science in the level of theoretical and practical knowledge. PO-2 On the basis of undergraduate level qualifications, develop and deepen the same or a different areas of information at the level of expertise, and analyze and interpret by using statistical methods PO-3 Develop new strategic approaches for the solution of complex problems encountered in applications related to the field and unforeseen and take responsibility for the solution. PO-4 Evaluate critically skills acquired in the field of information in the level of expertise and assess the learning guides. PO-5 Transfer current developments in the field and their work to the groups inside and outside the area supporting with quantitative and qualitative datas as written, verbal and visual by a systematic way. PO-6 Use information and communication technologies with computer software in advanced level. PO-7 Develop efficient algorithms by modeling problems faced in the field and solve such problems by using actual programming languages. PO-8 Respect to social, scientific, cultural and ethical values at the stages of data collection related to the field, interpretation, and implementation. PO-9 To solve problems related to the field, establish functional interacts by using strategic decision making processes. PO-10 Establish and discuss in written, oral and visual communication in an advanced level by using at least one foreign language.

Learning Outcomes LO-1 The student obtains a basic understanding of machine learning approaches and neural networks. LO-2 The student distinguishes the different categories of Machine Learning techniques and identify situations in which they might be used. LO-3 The student discusses the relationships between training set, test set, generalisation, cross validation. LO-4 The student describes how various machine learning techniques work and what their strengths and limitations are. LO-5 The student selects and use appropriate machine learning techniques to solve real problems.

Course Assessment Matrix:

	PO 1	PO 2	PO 3	PO 4	PO 5	PO 6	PO 7	PO 8	PO 9	PO 10
LO 1
LO 2
LO 3
LO 4
LO 5