GNR638 - Machine Learning for Remote Sensing II
The course is chill, mainly due to the chill evaluation scheme.
Time Commitment Required
5-6 hours per week
Grading Policy and Statistics
No hard prerequisite. Basic knowledge of Linear Algebra is expected. Having some very intro level ML familiarity (from random online course) will give you a slight advantage, and make the course chiller for you.
2 Paper Reviews - 25% (critical reviews of research papers)
1 Kaggle competition - 15%
Course Project - 35%
Final Quiz - 25%
Topics Covered in the Course
Recognition problems in computer vision and remote sensing, Evolution of feature extraction and representations, SIFT, Bag of Words, HOG, Neural Networks and CNNs, Loss Functions and Optimization, Discussion in detail of Important CNN architectures (VGGNet, AlexNet, GoogleNet, ResNets, etc.), advanced CNN models (Bayesian CNN, Siamese/Triplet CNN), Autoencoders, Object Detection approaches (versions of RCNN, YOLO, SSD), Image Segmentation, RNNs (Simple, LSTMs, GRUs), Attention, GANs, VAEs, Multi-modal learning, etc.
The professor has regular lectures where new concepts are covered. A lot of content is covered in a lecture, and depth is not a lot. The professor posts a lot of good optional reading material so students can explore interesting topics in greater detail on their own.
Because the course takes on a more overview type of approach, the content is more breadth over depth, so rigorous mathematical discussion and intuition of various ML approaches is lacking, and left as an exercise to the student.
About the teaching style, the professor will just come, deliver the content planned and leave without much interaction, so the students have to be proactive and ask doubts wherever required by interrupting the professor (he doesn’t mind at all)
Feedback on Exams
The final quiz is moderate in difficulty. The questions are mostly either conceptual, or right out of the slides. There are no questions from topics given for self reading. The exam format for us was MCQs or fill in the blanks with justifications.
Motivation for taking this course
This course gives a high level overview of most of the hottest topics in ML, leaving the student to explore more based on their interests. After taking this course, you will be able to understand the basic premise of most ML applications and approaches, and will also be able to think of ML-driven solutions to problems. Also, the evaluation scheme is very chill, and the grading is generous, so grading wise also the course is an ideal Institute Elective.
Some pretty niche and advanced ML topics are touched upon in the course, in a manner appreciable by ML novices too.
This course does a good job of giving a lot into the very broad and complex field of ML, and is able to demonstrate the myriad of applications of Machine Learning to real life problems.
How strongly would I recommend this course?
This course is like a Crash Course in ML, so you will have a good level of familiarity with all kinds of ML most people talk about, and will be able to implement a lot of simple models on all kinds of clean tasks. But, don’t expect a good deep understanding of the why and the how of the kinds of ML you cover in this course, and you will have to taking more focused courses if you want that. So this course is recommended for people who want a tour of ML without getting into the math, and for people sitting for placements who don’t have too much time to devote to learning ML but want to know enough of it to get through tests and interviews.
When to take this course?
I took this course in my 5th semester. This course can also be taken by sophies in the third semester, after which you can explore ML in a more rigorous manner through courses from the CS and IE departments, and through projects under professors.
Going forward, you can take more focused and rigorous Intro level course in ML like CS419 or EE769 (both recommended over DS303). You can also consider the many advanced ML and DL courses from the CS and IE departments. There is also the newly introduced AI&DS minor by CMINDS that is very flexible and easy to do in parallel with another minor. Also, there are a lot of professors working on some cool areas in machine learning so you can also consider doing some research under their guidance.
GNR 638 Review By: Shubham Lohiya