Instructor

Biplab Banerjee

Semester

Spring’20

Course Difficulty

Course content was easy to moderate if you have done Machine Learning before in some form. It was moderate to tough if this is your introduction to machine learning.

Time Commitment Required

Attending the lectures and revision before the exam was enough for the quizzes. The homework and assignments were a bit lengthy and required more time.

Grading Policy and Statistics

Relative grading was followed. The grading was pretty generous.
AA : 42
AB : 70
AP : 3
BB : 27
BC : 7
CC : 4
CD : 3

Attendance Policy

The professor had mentioned an attendance policy in the beginning of the course but did not implement it in the online semester.

Pre-requisites

Knowing python and experience of working with databases would provide an advantage as the assignments require dealing with data before applying machine learning algorithms.

Evaluation Scheme

3 Quizzes : 60%
Assignment : 20%
Project : 20%

Topics Covered in the Course

1. Various Supervised and Unsupervised Machine Learning algorithm ( Decision Trees, Linear Regression, Logistic Regression, KNN, Probabilistic Classifiers , SVM, Clustering)
2. Ensemble Learning
3. Deep Learning

Teaching Style

Professor primarily used slides for teaching and uploaded them regularly. Coding sessions were conducted weekly to help with the implementation on the algorithms learnt in class.

Tutorials/Assignments/Projects

Assignments and homework were lengthy and time taking. The project was open ended and involved reading up on literature and implementing ML algorithms.

Feedback on Exams

3 Quizzes were conducted throughout and midsem and endsem were not conducted. The quizzes were objective, moderate in difficulty and focused more on application of concepts learnt rather than mathematical aspect.

Motivation for taking this course

This is a compulsory course in the basket for DS Minor.

Course Importance

Gaining exposure to multiple machine learning algorithms increases your range for application in projects and you get a basic idea on which you can build with future courses.

How strongly would I recommend this course?

I would recommend this course to anyone who wants to explore the field of Machine Learning . The course is not very deep but covers a wide range of topics in ML.

When to take this course?

I took this course in my 4th semester. It is better to take this course after completing DS203 as you get familiar with handling data in python.

References Used

Mathematics for ML : https://gwthomas.github.io/docs/math4ml.pdf
Machine Learning : http://noiselab.ucsd.edu/ECE228/Murphy_Machine_Learning.pdf

DS 303 Review By: Varun Pathak