ADVANCED MACHINE LEARNING

AI436

Course Code: AI436
Semester: -
Theory+Practice: 3 + 0
ECTS: 5

Course Type

Elective Course

Course Level

Bachelor's Degree

Year of Study

Semester

ECTS

Name of Lecturer(s)

Learning Outcomes

At the end of this course, the students;

Know and apply evaluation methods.
Know how to design hybrid learning systems.
Know how to apply techniques to specific problems.

Mode of Delivery

Face to face

Prerequisites

Recommended Optional Programme Components

Course Description

Model selection and evaluation techniques, Ensemble methods: bagging, boosting, stacking, Regularization and sparsity (L1/L2 norms), Kernel methods and support vector machines in depth, Dimensionality reduction: PCA, t-SNE, UMAP, Probabilistic graphical models: Bayesian networks, Markov random fields, Generative models: Gaussian Mixture Models, HMMs, Variational Autoencoders (VAE), GANs, Deep learning architectures: CNN, RNN, LSTM, Transfer learning and domain adaptation, Semi-supervised and self-supervised learning, Model explainability and fairness in machine learning, Case studies in computer vision, NLP, and time series.

Course Contents

WEEK	TOPICS
1^st Week	Machine learning paradigms.
2^nd Week	Evaluation methods for machine learning.
3^rd Week	Reinforcement learning.
4^th Week	Cost-sensitive learning.
5^th Week	Analytical learning.
6^th Week	Combining inductive and analytical learning methods.
7^th Week	Active learning.
8^th Week	Mid-term
9^th Week	Applications to selected problems.
10^th Week	Applications to selected problems.
11^th Week	Applications to selected problems.
12^th Week	Applications to selected problems.
13^th Week	Applications to selected problems.
14^th Week	Applications to selected problems.

Recommended or Required Reading

* Pattern Recognition and Machine Learning - Christopher M. Bishop
* Deep Learning - Ian Goodfellow, Yoshua Bengio, Aaron Courville

Planned Learning Activities and Teaching Methods

Lecture
Questions/Answers
Practice
Problem Solving
Project
Report Preparation
Presentation

Assessment Methods and Criteria

	Quantity	Percentage (%)
Mid-term	1	30
Project	1	30
Total(%)		60
Contribution of In-term Studies to Overall Grade(%)		60
Contribution of Final Examination to Overall Grade(%)		40
Total(%)		100

Language of Instruction

Turkish

Work Placement(s)

Programme Qualifications / Learning Outcomes

	LO1	LO2	LO3
K1
K2
K3
K4
K5
K6
K7
K8
K9
K10
K11
K12