TÜRKÇE
ENGLISH
Information
| Unit | INSTITUTE OF NATURAL AND APPLIED SCIENCES |
| ELECTRICAL-ELECTRONICS ENGINEERING (PhD) (ENGLISH) | |
| Code | EE726 |
| Name | İstatistiksel Makine Öğrenimi |
| Term | 2026-2027 Academic Year |
| Term | Spring |
| Duration (T+A) | 3-0 (T-A) (17 Week) |
| ECTS | 6 ECTS |
| National Credit | 3 National Credit |
| Teaching Language | Türkçe |
| Level | Belirsiz |
| Type | Normal |
| Mode of study | Yüz Yüze Öğretim |
| Catalog Information Coordinator | Doç. Dr. FATİH KILIÇ |
| Course Instructor |
The current term course schedule has not been prepared yet.
|
Course Goal / Objective
The objective of this course is to develop a deep understanding of statistical machine learning and to provide students with comprehensive knowledge and skills in this field. The course is designed to provide students with the statistical foundations of modern machine learning methods, to develop their ability to analyze complex data sets and to conduct original research using statistical modeling techniques.
Course Content
This course focuses on foundational statistical principles, providing students with the statistical fundamentals of machine learning methods. The course aims to equip students with the skills to understand, apply, and evaluate machine learning algorithms on diverse datasets. Additionally, it covers advanced topics offering students a comprehensive knowledge base in the field of statistical machine learning.
Course Precondition
There is no prerequisite for the course.
Resources
1. Machine Learning - A Probabilistic Perspective, Kevin P. Murphy, MIT Press, 2012. 2. Machine Learning with PyTorch and Scikit-Learn: Develop machine learning and deep learning models with Python by Sebastian Raschka, Yuxi (Hayden) Liu , Vahid Mirjalili, 2022. 3. Introduction to Statistical Learning, Python Edition, By Bala Priya C, KDnuggets on July 28, 2023 in Python
Notes
Dangeti, P. (2017). Statistics for machine learning. Packt Publishing Ltd.
Course Learning Outcomes
| Order | Course Learning Outcomes |
|---|---|
| LO01 | Explains the statistical and probabilistic foundations of machine learning (probability theory, statistical inference, optimization, linear algebra), analyzes the role of these foundations in learning algorithms, and applies them to appropriate problems. |
| LO02 | Comprehends supervised and unsupervised learning methods (regression, classification, regularization, kernel methods, clustering, dimensionality reduction) along with their theoretical assumptions, and develops models on real-world data. |
| LO03 | Evaluates the performance of developed models using appropriate statistical metrics (cross-validation, ROC/AUC, bias-variance decomposition), and critically applies the concepts of overfitting and generalization to model selection and hyperparameter optimization processes. |
| LO04 | Designs and optimizes models for complex problems by integrating deep learning architectures (multi-layer networks, convolutional and recurrent networks, attention mechanisms, transformer structures) with modern statistical learning approaches. |
| LO05 | Critically evaluates recent machine learning approaches (self-supervised learning, transfer learning, Bayesian deep learning) by following current scientific literature, and addresses ethical issues in applications (data privacy, algorithmic bias, explainability) within the framework of responsible AI principles. |
Relation with Program Learning Outcome
| Order | Type | Program Learning Outcomes | Level |
|---|---|---|---|
| PLO01 | Bilgi - Kuramsal, Olgusal | Being able to specialize in at least one of the branches that form the foundations of Electrical and Electronics Engineering by increasing the level of knowledge beyond the master's level | 4 |
| PLO02 | Bilgi - Kuramsal, Olgusal | To comprehend the integrity of all the subjects included in the field of specialization. | 3 |
| PLO03 | Bilgi - Kuramsal, Olgusal | Having knowledge of the current scientific literature in the field of specialization to analyze the literature critically | 4 |
| PLO04 | Bilgi - Kuramsal, Olgusal | To comprehend the interdisciplinary interaction of the field with other related branches, to suggest similar interactions. | 3 |
| PLO05 | Bilgi - Kuramsal, Olgusal | Ability to do theoretical and experimental work | 2 |
| PLO06 | Bilgi - Kuramsal, Olgusal | To create a complete scientific text by compiling the information obtained from the research | |
| PLO07 | Bilgi - Kuramsal, Olgusal | To work on the thesis topic programmatically, following the logical integrity required by the subject within the framework determined by the advisor. | |
| PLO08 | Bilgi - Kuramsal, Olgusal | To search for literature in scientific databases, particularly the ability to correctly and accurately scan databases and evaluate and categorize listed items. | 4 |
| PLO09 | Bilgi - Kuramsal, Olgusal | Having a command of English and related English jargon at a level that can easily read and understand a scientific text written in English in the field of specialization and write a similar text | |
| PLO10 | Bilgi - Kuramsal, Olgusal | Ability to write a computer program in a familiar programming language, generally for a specific purpose, specifically related to the field of expertise. | 4 |
| PLO11 | Bilgi - Kuramsal, Olgusal | Ability to plan and teach lessons related to the field of specialization or related fields | |
| PLO12 | Bilgi - Kuramsal, Olgusal | Being able to guide and take the initiative in environments that require solving problems related to the field | 3 |
| PLO13 | Yetkinlikler - Bağımsız Çalışabilme ve Sorumluluk Alabilme Yetkinliği | Ability to communicate with people in an appropriate language | |
| PLO14 | Yetkinlikler - Öğrenme Yetkinliği | Adopting the ethical values required by both education and research aspects of academician | |
| PLO15 | Yetkinlikler - Öğrenme Yetkinliği | To be able to produce projects, policies, and processes in the field of expertise and to evaluate these elements | 3 |
| PLO16 | Yetkinlikler - Öğrenme Yetkinliği | Ability to research new topics based on existing research experience | 3 |
Week Plan
| Week | Topic | Preparation | Methods |
|---|---|---|---|
| 1 | Overview of machine learning and its statistical foundations, basic concepts in probability theory and statistical inference | Olasılık dağılımları, beklenen değer ve varyans kavramlarını tekrar edin. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 2 | Linear regression and multiple linear regression, model evaluation and performance metrics | Study assumptions of simple and multiple linear regression. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 3 | Regularization techniques (L1 and L2 regularization) | Review overfitting and bias-variance tradeoff. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 4 | Polynomial regression and Logistic regression | Learn classification problems and sigmoid function. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 5 | Decision trees and ensemble methods (Random Forests, Gradient Boosting) | Study decision trees and ensemble learning concepts. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 6 | Support Vector Machines (SVM): Classification | Learn margin, kernels, and linear separability. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 7 | Support Vector Machines (SVM): Regression | Study SVR and epsilon-insensitive loss. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 8 | Mid-Term Exam | konularını tekrar edin. Review first 7 weeks. | Ölçme Yöntemleri: Yazılı Sınav |
| 9 | K-means clustering, hierarchical clustering methods, evaluation of clustering algorithms | Study K-means and distance metrics. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 10 | Dimensionality Reduction, principal component analysis (PCA), t-Distributed Stochastic Neighbor Embedding (t-SNE) | Learn PCA and variance explanation concept. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 11 | Bayesian methods in machine learning | Review Bayes theorem and prior-posterior relation. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 12 | Deep Learning: Neural network architecture and activation functions | Study basic components of neural networks. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 13 | Training neural networks: backpropagation and optimization techniques | Review gradient descent and derivatives. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 14 | Recent advancements and trends in statistical machine learning | Explore recent ML papers and trends. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 15 | Bayesian analysis | Study Bayesian inference and probabilistic modeling. | Öğretim Yöntemleri: Anlatım, Soru-Cevap, Tartışma |
| 16 | Term Exams | Review all topics. | Ölçme Yöntemleri: Yazılı Sınav |
| 17 | Term Exams | tamamlayın ve tekrar yapın. Complete missing topics and revise. | Ölçme Yöntemleri: Yazılı Sınav |
Student Workload - ECTS
| Works | Number | Time (Hour) | Workload (Hour) |
|---|---|---|---|
| Course Related Works | |||
| Class Time (Exam weeks are excluded) | 14 | 3 | 42 |
| Out of Class Study (Preliminary Work, Practice) | 16 | 5 | 80 |
| Assesment Related Works | |||
| Homeworks, Projects, Others | 1 | 24 | 24 |
| Mid-term Exams (Written, Oral, etc.) | 1 | 2 | 2 |
| Final Exam | 1 | 2 | 2 |
| Total Workload (Hour) | 150 | ||
| Total Workload / 25 (h) | 6,00 | ||
| ECTS | 6 ECTS | ||