Grad Coursework @ ECE, Virginia Tech!

So yeah! Last few days I have been asked quite a lot of questions on the topic of basically deciding which admits to choose while deciding where to join for a Master’s degree. Quite a few people have then also asked me about the courses at VT(Virginia Tech). I have also been in the same shoes and have pestered quite a few seniors with a lot of super specific questions regarding deciding which courses to enroll!

Based on my experience,I know that coursework is an important part of graduate education. Hence, they should play a critical role in deciding where to start your graduate education.

Hence, I am writing this blog as my opinion of the courses I took as a Master’s student in the Electrical and Computer Engineering Department at Virginia Tech (Go Hokies!).

Fall 2017

ECE 5554 : Computer Vision

This is an excellent introduction to computer vision course taught by Prof Dr. Jia Bin-Huang. The course is neatly structured and I would definitely recommend going through the course website before you choose to enroll. I really liked this course because of the hands-on programming assignments in MATLAB, which are, implementing the concepts learned in the class during the week. Though, a fair warning: The homeworks are pretty challenging because they usually take up a lot of time to code-up from scratch. The difficultly level of the homeworks is not impossible, but they are definitely time-consuming. Additionally, the course also requires a project. The professor has high expectation when it comes to the project but the grading is pretty lenient.

Pros : Excellent introduction to computer vision, Cool Project to showcase on your resume, Loads of coding in MATLAB

Cons : Not a lot of time spent on Deep Learning, Homeworks take too much time

Pro-Tip : Go through the course-website, it has everything you need before and after you decide to take this course

STAT 5525 : Data Analytics 1

I would say this was one of the best courses I have ever attended. The course was taught by Prof Dr. Scot Leman. This course introduces you to basic concepts of Machine Learning/ Data Analytics(from a stats perspective). The Professor is an excellent teacher. He literally walks you through the complex math behind various statistical methods such as Regression Trees, Logistic Regression, Naive Bayes and other such concepts. His homeworks are simple and thoroughly help you understand the concept. He has a semester long project/competition wherein he teams the class up in groups and gives them the same data-set from which you are supposed to present some suspicious findings! It is an extremely cool assignment as you get to experiment and play around with the data-set which helps you gain the knack for solving different data analytics problems. The other reason why I liked the course is because the professor keeps the class constantly engaged(Note: Class used to be @ 8 AM) with his anecdotes and different stories.

Pros : Excellent Introduction to Machine Learning (especially if you are not ready to face the crazy math behind it on your own as taught in the ECE — Advanced Machine Learning by Prof Dr. Bert Huang), Hands-on Python/MATLAB (depending how you solve your assignments), $200 gift card if you win the competition

Cons : This course is not being taught by Prof Leman for Fall 2018 (Advanced Machine Learning is being taught by Prof Plassmann (might be easier than the last time), Data Analytics will be taken by a prof in the CS department.)

Pro-Tip : Definitely take it if taught by Prof Leman! Have a look at the course website here (Doesnt contain much info, as the Prof generally used to chalkboards to teach and not PPTs)

ECE 5565 : Network Analysis and Protocols

This is the super fundamental course which introduces TCP/IP model and other the basic networking topics to students. The course closely follows the Kurose-Ross book : Computer Networking — A Top Down Approach. The course is taught by Prof. Dr Yaling Yang. The course material is mainly the PPTs that are derived from the book and is fairly straight forward. The homework assignments are ridiculously easy. The programming assignments are fairly simple where you have to build a UDP client-server. Some of the homework involves going through the TCP/UDP/IP packets using Wireshark and determining various fields in them.

Overall, I would say if you have taken an undergrad course which covers TCP/IP and UDP, then please do NOT take this course. The grading is lenient except for the finals which might end up messing your grade (as it did for me!)

Pros : Super Easy Course (can be used to balance out other difficult ones), Intro to Networks, easy A (if you don’t mess up the finals)

Cons : Super Easy Course (Not learning anything new, hence in my opinion waste of time, energy and money), Final Exam grading is tough, No Projects

Pro-tip : Don’t take it if you don’t intend to pursue networks

So those were the courses that I took for Fall 2017 and some other courses that I would recommend are Hardware Software CoDesign, Advanced Topics in Decision Making, Advanced Machine Learning.

So, I hope this was helpful and I will probably write another post about the courses I completed in Spring 2018 later.

Shoot me an email, if you have any additional queries regarding deciding courses I could definitely help you out or at least try to point you in the right direction.