Sem1 Update!
Semester 1 Update - Stony Brook University Fall 19 MS CS!
Today marks 3 months since I landed in New York in August. It’s been a roller-coaster ride so far at SBU both from an academic and personal standpoint. I have learned to cook - I can get away with making a sandwich and simple dal-chawal (lentils and rice) and I am pretty decent at that too! :) Also, I can make them in under 30 minutes on a good day!
You may say, why the hurry? Take your time and make a nice meal for yourself! This brings me to the academic life of a Computer Science Graduate student at SBU. It is hectic, to say the least. There have been weeks where I slept just 3 hours on an average daily. Four graduate-level courses is not a joke! Do not underestimate the coursework, the assignments, and the exam prep.
If you were one of the smart ones who stuck to light courses just to keep your life easy and have a great GPA at the end of the Fall semester - more power to you! Unfortunately, I did not come all the way here, spend thousands of dollars just so I could breeze through it. I came here to study and study I shall! I have taken up courses I wanted to study, to explore and hopefully make a career out of. I know they are tough. I know that I might need to sacrifice a few hours of sleep, but I hope that this will all pay off, if not now, maybe sometime in the future!
I end my rant here, and focus on the good stuff going forward!
After, working for 4 years in the Software Engineering domain and having only tasted the world of Machine Learning and all its applications in other fields I knew I had to take related courses. I would mark the time when I took the online course for Mathematics for Machine Learning by Imperial College, London, on Coursera as the pivot to my journey and growing fondness towards Machine Learning and its applications. This pushed me to take courses for Data Science Fundamentals, Natural Language Processing, and Computer Vision. The fourth course I am enrolled in is Analysis of Algorithms - more of a refresher of what I studied during my undergraduate education.
Below is a review of the things I have learned in each subject in brief for each course. I hope to write more in detail blog posts about specific topics that I have found fascinating, especially in Computer Vision and Natural Language Processing.
Data Science Fundamentals (CSE-519):
The course is infamously construed as light but I don’t think it is. It requires significant effort on your part, especially in the assignments (which are participation in Kaggle challenges) and a lot of Exploratory Data Analysis (EDA) to unearth interesting facts in the given dataset. Finding something interesting is one thing, conveying it through the means of a chart/graph is an altogether different beast. (Hopefully, the Visualization course offered next semester talks more about it).
Frameworks/Tools/Technologies/Buzz Words: Pandas, Numpy, Scikit-Learn, Matplotlib, Jupyter Notebooks, Google Colab, Regression, Classification, etc.
Introduction to Computer Vision (CSE-527):
This is a course I would recommend to everyone who ever joins Stony Brook as a Grad student. The primary reason it has some of the best assignments for learning the material! A lot of this reflects on the TAs who have managed this course particularly well. This is definitely a course that demands a lot from you. The current assignment I am working on is not just forcing me to think about the core homework material but also forcing me to work in difficult restricted environments because of limited computing resources, limiting accesses to files on disk, being judicious with memory on the GPU, etc.
Frameworks/Tools/Technologies/Buzz Words: Most things from DSF above, OpenCV, Tensorflow, Pytorch, Convolutional Neural Networks (CNNs), Generative Adversarial Networks (GANs), etc.
Natural Language Processing (CSE-538):
This is again a course I would recommend. The course should be renamed to NLP using Deep Learning as it mostly concentrates on neural-network based approaches. The first few lectures do go through some topics like Naive Bayes Classifier and Logistic Regression but after that, it is neural all the way! It was not managed in the best possible way, but the assignments are again really good.
*Frameworks/Tools/Technologies/Buzz Words: All assignments use Tensorflow. CNNs, Recurrent Neural Networks(RNNs), Long Short Term Memory (LSTMs), Deep Averaging Networks (DANs), Gated Recurrent Units (GRU), Transformers, Dependency Parsing, etc. *
Analysis of Algorithms (CSE-548):
This is an Algorithms course just like one would expect - textbook-style! A good course for someone who has lost touch a bit (like me!) to refresh many basic concepts right from sorting to greedy algorithms, dynamic programming and so on. The professor is punctual and has really managed the course very well. My only complaint here is that the assignments do not reflect the questions one might see on the exam! Assignments are fairly basic and easy but time-consuming and do not do much in terms of preparing you for what lies ahead in exams or towards internship/job prospects.
Apart from all of this, I am also a Teaching Assitant for an undergraduate course, Programming Abstractions. This has been easy work so far and I picked up on some OCaml along the way. I got a chance to write scripts to automate homework submissions of students as well!
It feels good to be back in school after my stint in the industry! So much to learn and so much to explore! I need to head back to my Computer Vision assignment that has been staring at me all this time that I spent writing this! But I committed to writing this piece this weekend and even though it is technically Monday, like my friend Aman says, it isn’t the next day until you’ve gone to bed! - This is wrong on so many levels but I will leave that discussion for another time!
The perfectionist in me says not to post this as I am sure I can improve on this and write more clearly, better grammar and vocabulary or more details on the courses. The realist in me says that I might never get to that and its best to post this as is - raw, straight from the works in the brain! If you read this, you know I chose the realist in me!
If you want to know more about the life of a grad student in CS at SBU or just want to chat and know more about me or have opportunities for me - hit me up on LinkedIn or any one of the social network options on the left here (or on top if you are on mobile).
Peace!