CSE3220 - Machine Learning (Fall, 2025)

Course Time:	Thursday- Formal Session:13:30-15:30 / 19:00-21:00 (Single 10 minute break)
Place:	C202 / C204
Instructor:	Prof.Dr. Muhammet G. C. ERDEM, muhammet.gokhan.erdem@{gmail, yahoo}.com
Assistant(s):	-------
Class Group	Computer Engineering ML Class Google Group / MLClass@classroom.google.com
Description:	Machine Learning Introduction, Regression, Multivariate Regression, Gradient Descent Learning, Logistic Regression, Regularization, Artficial neural networks, biological foundations, cognitive processes and their modeling using artificial neural networks, problem solving using neural networks, supervised/unsupervised learning, Support Vector Machines, Principle Component Analysis, Naive Bayesian Learning.
Prerequisites:	Python or Matlab or one of the Programing languages (Java, C#), Statistics & Mathematics background for engineering departments
Textbooks:	2021, Zhi-Hu Zhou, “Machine Learning”, Springer, ISBN: 9789811519673 2021, Chris Mattmann, “Machine Learning with Tensorflow “, Manning, ISBN: 9781617297717 2021, Francois Chollet, “Deep Learning with Python”, Manning Publishing, ISBN: 9781617294433 2018, Birol Kuyumcu, “OpenCv Goruntuleme ve Yapay Ogrenme”, Level Kitap,  ISBN: 9786056567933 2017, Sebastian Raschka, “Python Machine Learning 2.nd edt”, PACKT Publishing, ISBN: 978-1-783-55513-0 2014, Ethem Alpaydin, “Introduction to Machine Learning 3rd edt”, MIT Press, ISBN: 978-0-262-02818-9 2012, Kevin P. Murphy, “Machine Learning”, MIT Press, ISBN: 978-0-262-01802-9 2009, Simon Haykin, “Neural Networks and Learning Machines - 3rd edt” Prentice Hall, ISBN: 978-0-131-47139-9 2006, Christopher Bishop, “Pattern Recognition and Machine Learning”, Springer, ISBN: 978-0-387-31073-2 2004, Mehmet Onder Efe, “Artificial Neural Networks and Their Applications”, Bogazici University Publications, ISBN: 975-5-18223-3 1994, Laurene Fausett, “Fundementals of Neural Networks”, Prentice Hall, ISBN: 978-0-133-34186-7. Free Machine Learning with Matlab
Projects	All Projects are due at the beginning of class. Due dates for projects will be announced at least a week ahead of time. No late submission will be accepted!.
Tests:	Students will have one final exam. Students have to submit their final Technical Report to any journal in SCI or SCI-Expanded or conference in related area.
Grading:	Choice1: %30 MidTerm + %20 Homeworks + %25 FinalExam + %25 Final Project Choice2: %40 MidTerm + %60 FinalExam Choice3: %40 MidTerm Hackathon + %60 FinalExam Hackathon Choice4: %40 Hybrid MidTerm + %60 Hybrid FinalExam
Late Submission:	Late submission of projects will not be accepted.
Programming Platforms/Tools:	1. Python 3.5 (or newer versions) + Machine Learning Libraries 2. Matlab and Simulink Bundled Student Edition 3. Visiual Studio 2015 (or newer)  + {OpenCV / Accord.NET etc..} 4. Visual Code
Suplimentary Best Video Lectures:	1. Andrew NG - Machine Learning - Stanford 2. Yaser Abu Mostafa - Machine Learning - Caltech 3. Patrick Winston - Artificial Intelligence - MIT 4. Emily Fox - Machine Learning

Weekly Based Lectures:

Lecture01 - Introduction to Machine Learning
Lecture02 - Linear Regression with One Variable
Lecture03 - Linear Regression with Multiple Variables
Lecture04 – Regularization (Ridge & Lasso Regressions)
Lecture05 - Logistic Regression
Lecture06 - SVM
Lecture07 – PCA (Dimension Reduction)
Lecture08 - Neural Network Fundementals, Components
* Prof. Leslie Smith - Introduction to Neural Network
* Christos Stergiou and Dimitrios Siganos - Introduction to Neural Networks
* Genevieve Orr - Motivation for Neural Networks
* Learning Artificial Neural Networks [Robotics]
* Perceptron - Applet Demo
Lecture09 - Neural Network Models
Lecture10 – Clustering (SOM, LVQx)
Lecture11 - NeoCognitron - “Convolutional Neural Net”(CNN)
Lecture12 – CNN Based Models
Lecture13 – Exampler Domain Apps: Speech Recognition
* AquaPhoenix - Audio Processing with Matlab
* Bern Plannerer - Speech Recognition with Matlab
* Ian McLoughlin - Applied Speech and Audio Processing with Matlab
* David Roberts - Voice Recognition Project