Machine Learning

Theory & Lab

Course Description

The Machine Learning course provides a comprehensive exploration of the principles, algorithms, and applications that underpin the field of machine learning. The curriculum covers fundamental concepts such as supervised, unsupervised, and reinforcement learning, as well as advanced topics like deep learning and natural language processing.

Through a combination of theoretical lectures, practical exercises, and hands-on projects, students will develop a deep understanding of various machine learning techniques and their real-world applications. The course emphasizes the theoretical foundations of machine learning algorithms, teaching students how to preprocess data, select appropriate algorithms, and evaluate their performance.


Course Rationale

The science of getting computers to act without being explicitly programmed, machine learning has given us self-driving cars, practical speech recognition, efficient web search, and a vastly improved understanding of the human genome. In this class, you will learn about the most effective machine-learning techniques and practice putting them to use. You will not only learn about the theoretical underpinnings of learning but also gain the practical knowledge required to quickly and powerfully apply these techniques.


Contact Information

Instructor: Md. Riad Hassan
Office: A-510
Email: riad@cse.green.edu.bd


Class Schedule

Day Time Room
Saturday 01:30 PM - 03:00 PM K104
Sunday 01:30 PM - 03:00 PM K104
Monday 01:30 PM - 03:00 PM K104

Counseling Hours

Day Time
Saturday 11:30 AM - 01:00 PM
Sunday 11:30 AM - 01:00 PM

Course Outcomes (COs)

Upon successful completion of this course, students will be able to:

  • CO1: Understand and apply fundamental machine learning concepts, such as supervised, unsupervised, reinforcement learning, online learning, and Bayesian rules.
  • CO2: Acquire critical thinking and problem-solving ability via analyzing data, identifying patterns, and formulating effective ML models.
  • CO3: Comprehend the outputs of machine learning techniques, fostering a greater willingness to explore advanced topics in the field.

Topic Outline

Lecture Topic
1 Introduction
2-6 Supervised Learning (Regression, Classification, Linear/Logistic Regression, Cost Functions, Optimization) - Assignment 1
7-10 Bayesian Decision Theory (Probability, Uncertainty, Likelihood, Posterior, Naive Bayes)
11-12 Parametric and Non-parametric Methods for Density Estimation - Quiz 1
13-14 Unsupervised Learning (Association Rules, K-Means Clustering)
15 Midterm Exam
16-19 Perceptron & Multilayer Perceptrons (ANN Architecture, Cost Function, Backpropagation, SGD, Hyperparameter Tuning) - Group Project
20-21 Introduction to Graphical Models - Quiz 2
22-25 Time Series Modeling / Online Learning (Markov Models, Hidden Markov Models, Bayesian Networks)
26-28 Reinforcement Learning (Markov Decision Processes, Q-learning)
29-30 Design and Analysis of Machine Learning Experiments
  Final Exam

Reference Materials


Assessment Methods

The final grade will be calculated based on the following components:

Assessment Method CO1 CO2 CO3 Total
Final Exam 20% 20%   40%
Midterm Exam 15% 15%   30%
Class Tests (Best 2 out of 3) 5% 5%   10%
K/S/A Test 1 (Attendance + Assignment)     10% 10%
K/S/A Test 2 (Presentation)     10% 10%
Total 40% 40% 20% 100%

Course Policies

  • Assignments: Late submissions will receive a zero mark.
  • Class Tests: At least three tests will be held; the best two will be counted. No makeup tests.
  • Examinations: Closed book, closed notes. Mobile phones are strictly prohibited.
  • Devices: The use of laptops and mobile devices during exam is strictly discouraged.