COMP562 Intro Machine Learning

Table of Contents

Course information #

Machine learning as applied to speech recognition, tracking, collaborative filtering and recommendation systems. Classification, regression, support vector machines, hidden Markov models, principal component analysis, and deep learning.

I took this course under Prof Jorge Silva in Spring 2022

Schedule #

Tuesday & Thursdays

Time slot: 5:00pm - 6:15pm

Course breakdown #

35% Assignments
20% Midterm exam.
30% Final exam.
15% Final project.

The exams were not as hard as the contents of the lecture. Studying the slides and the notes will really help in the exams.

Prerequisites #

Prerequisites:

(COMP 211 and 301) or (COMP 401 and 410)
MATH 233, 347, and STOR 435

A grade of C or better is required in all prerequisite courses. Permission of the instructor for students lacking the prerequisites.

Course details #

These are the following topics which were tested for the final exams.

Probability Distributions and Optimization
Linear Models for Regression
Linear Models for classification
Directed Graphical Models
Mixture Models and EM
Latent Linear Models
Sparse Linear Models
Kernel Methods
Neural Networks and Deep learning

Assignments #

The written assignments will contain questions based on the material covered in the lectures. Some assignments may contain questions from the required textbook. Assignments will be posted on Sakai. Once posted, assignments will typically be due in 1 or 2 weeks as specified on the assignment document. There will be 4 assignments during the semester.

Ratings #

Workload 3/10 #

Personally, the workload for the course is not very high to do ok for the course. However, to fully understand all the contents within the course, an exponentially higher amount of effort is required. (10/10 high workload if you want to understand everything)

Organization 7/10 #

The course starts of relatively easier at the start and increases in difficulty towards the end.

Learning 5/10 #

Due to my lack of understanding of the course during the lectures, I did not learn too much from the course.

Enjoyment 2/10 #

Personally, I did not enjoy doing the course. It was very math and statistics heavy.

Usefulness 8/10 #

When learnt in its full glory, this course is very useful.

Overall 5/10 #

Overall I did not enjoy the course but it will be very useful for anyone who is looking to do machine learning in the future.

Final Grade: A-