COURSE DESCRIPTION: Fundamentals of machine learning and pattern recognition. Topics covered include neural networks, Bayesian inference, and non-parametric techniques.
COURSE PREREQUISITES: DATA 601 and fundamental knowledge of MATLAB.
TEXTBOOK (required):
- Introduction to Machine Learning by Ethem Alpaydin, 4th ed (2020)
Recommended Books:
- Pattern Recognition by S. Theodoridis and K. Koutroumbas.
- Pattern Classification, by R. O. Duda, P. E. Hart, D. G. Stork.
METHODOLOGY: Lectures, course term project, tests, programming assignment(s), possible pop quizzes; all online. This course will be primarily delivered with synchronous lectures. Some programming application assignments will be included to reinforce the concepts. A term project will be assigned to apply and reinforce the information covered in the class. Pop-up quizzes may be utilized. Tests will be given. Blackboard may be utilized for the delivery of some tests.
A Blackboard Course Shell will be set up and utilized for all communication, that is, live and recorded course presentations, HW/project assignments, course slides, notes, announcements, tests, quizzes, etc. You need to log in to the course shell daily. Also check the announcements section right before each class meeting starts, for any possible recent announcements pertaining to that day’s meeting.
Course Learning Outcomes:
- Understand the key concepts in machine learning
- Characterize the process to train and test machine learning algorithms and recognize ways to evaluate machine learning systems
- Apply machine learning systems to perform various artificial intelligence tasks by using a programming language such as MATLAB (preferred) and Python; experiment with the machine learning libraries of such languages
- Conduct and communicate data analysis/classification research, i.e. propose a novel research idea on a dataset, design and execute classification experiments to support the proposed idea, and write a report about the project and present it
Syllabus
Lecture 1: Introduction, the course, syllabus, website(s) for data. Probability and Random Variables Basics
Lecture 2: Ch. 2 Supervised Learning
Lecture 3: Ch. 3 Bayesian Decision Theory
Lecture 4: Ch. 4 Parametric Methods
Lecture 5: Ch. 5 Multivariate Methods
Lecture 6: Ch. 6 Dimensionality Reduction
Lecture 7: Ch. 7 Clustering
Lecture 8: Mid-term team project update presentations and Mid-term Test (online)
Lecture 9: Ch. 8 Nonparametric Methods and Ch. 9 Decision Trees
Lecture 10: Ch. 10 Linear Discrimination Analysis
Lecture 11: Ch. 11 Multilayer Perceptrons / ANNs
Lecture 12: Ch. 12 Deep Learning
Final Week: Final Test / Online (comprehensive)
Notes:
We will go over the entire textbook but some subsections may be skipped. So it is important that you attend and follow the class.
All lectures will be delivered synchronously, and they will be recorded for later availability in case you miss a lecture. If I won’t be able to do any synchronous lecture for some reason, or due to a conflict (e.g. if I have to attend a meeting, conference, etc.) I will upload a prerecorded lecture and I will notify you on Blackboard and I will ask you to view/listen to the prerecorded lecture within a certain amount of time.
Assignments, tests, presentations, quantity, scheduling, due dates, and dates may change. All dates are tentative. The instructor will do their best to stick to the above plan and notify students on Blackboard of any changes if any.
ATTENDANCE POLICY: Attendance to the synchronous lectures on Blackboard is expected and you should understand that participation is an essential element in the learning process. The instructor would appreciate an email message from students who are not going to attend any virtual class for whatever reason.
PARTICIPATION: Students are encouraged to ask questions and make comments, respond to the instructor’s questions, participate in the virtual discussions on the Blackboard Discussions/Forum items/threads; during the Q&A portion of the lecture sessions, and/or during the instructor’s virtual office hour sessions, and/or via email. Asking questions and making comments/suggestions to your peer’s presentations is also will be part of your participation grade.