Surviving Semester 1 At Stony Brook

Surviving Semester 1 at SBU

I just wrote my last final exam a couple of days ago and wanted to reflect on how it has been so far studying at SBU towards my Master’s degree. Hopefully , those thinking of applying to and incoming students at SBU can leave this blog with better information as to how they should approach their first semester from an academic viewpoint focused on getting good grades.

Before I rate the difficulty and hours involved I need to give you some context:

1. I am an average student, no genius here. People have managed better grades with a similar workload.
2. I enrolled for courses DSF, NLP, CV and AoA. This is NOT something I would recommend if you are an average person like me and wish to have top -notch grades. I chose these because I wanted to study these and having a high GPA was not a motivation for me (at least at the start of the semester).
3. You should note however that getting good grades is important as that can become a critical factor in many things right from finding an advisor for your advanced project (523/524) to finding a decent job post your studies . GPA might not matter so much on the latter but it definitely does not hurt . More importantly, if you wish to pursue a Ph.D. then this becomes all the more important to you.

I rate each course based on Difficulty and Learning Quotient on a scale of 10. I loosely define Learning Quotient as the ratio of what I think I learned to how much I expected to learn. Also, I add tentative amount of work involved each week.

Course Reviews In A Nutshell:

Data Science Fundamentals is true to its name - it is about basic concepts of a mixture of things from probability, statistics, linear algebra, some machine learning and a bit of big data. It does NOT go into the depths of any of these topics. If you’ve already done some Data Science then this is probably NOT the course for you. But, if you just want the good grades then go for it. Keep in mind that just because the course is labeled easy doesn’t mean there is no effort to be put in. It takes a good number of hours of work , week in week out to get good grades here. Do NOT mistake this course for an easy A! The concepts are easy but the work is not.

• Difficulty: 6/10
• Learning Quotient: 8/10
• Work Hours: 10+ hrs/week

Analysis of Algorithms - I took this under a Professor who had a really unbalanced grading scheme - 80% of your grade is decided based on 2 exams, a mid-term and a final. This is really stressful because if you do badly here then there is no saving you. The assignments are just too many questions but count for only 20% of the grade. I would recommend this course only if you are really really comfortable with Algorithms (Notice the 2 ‘really’ I used ). Do not treat this as a refresher course for your undergrad Algorithms course - This is a different beast!

• Difficulty: 12/10
• Learning Quotient: 8/10
• Work Hours: 5-10 hrs/week

Computer Vision - This is a cool course - I loved the assignments! They were well structured and the last couple also gave you some freedom to explore. Most assignments are more Deep Learning-based, so some exposure to it helps - but is not a pre-requisite. The exams are good too (I messed up mine) but I would highly recommend this course to any vision enthusiast. The assignments take a good amount of your time every week so start early and try to finish early.

• Difficulty: 9/10
• Learning Quotient: 9/10
• Work Hours: 10+ hrs/week

Natural Language Processing - Top drawer NLP course - should be renamed to NLP using Deep Learning in my opinion. Amazing assignments and an understanding instructor. Exams were good and the questions require you to think of applying the knowledge you learned in real-world scenarios. Slightly mismanaged but students were kept informed at all times for anything and everything. The class forum was highly interactive as well - which is a really good sign for any course!

• Difficulty: 9/10
• Learning Quotient: 10/10
• Work Hours: 10+ hrs/week

Points to keep in mind for your semester:

1. Stay on track at all times: Never let things go out of hand - it becomes really difficult once you fall back to come up to speed with everything. Start assignments early and wrap them up early. In courses that allow you to discuss - talk to others and bounce ideas off of each other and complete your assignments sooner. I cannot stress on this point enough - STAY ON TOP OF THINGS AT ALL TIMES! It might sound easy but is not so do not underestimate this rule!

2. Internship Hunt: The most important of this not how much you do once you are here but how much you can get done before you get here. You will be applying for internships during your first semester. Start applying early. Update your resume before you even arrive at SBU and keep updating it regularly. Get some competitive coding practice under your belt as well. Refresh on them Data Structures and Algorithms. Once you have an internship offer - it is just one less thing to worry about.

3. Want it easy?:

   • Do NOT take more than one heavy course! By heavy I mean anything that involves more than 10 hours of work every week should be hard.
   • See which courses do NOT have a final exam and if they have a decent grading scheme. Finals week can be stressful if you have too many exams back to back, so the lesser the better. While some information is available on Classie-Evals, your best bet would still be asking seniors around before finalizing courses.

RISKY TIP (Haven’t heard anybody do this - DO IT AT YOUR OWN RISK!):

You are required to have 12 credits as an international student in your first two semesters. If you can somehow manage to get a TA-ship for 2 credits (for a beginner level undergraduate course) and a 1 credit seminar then, you can just enroll in 3 courses and still reach that requirement for the first semester. This would be an ideal situation I feel for getting good grades as your workload is substantially reduced.

Ideally, it should never come to this - it feels like manipulating your interests to get good grades. One should ideally strive to study and the grades are what they are. But unfortunately, we live in a world where they look at grades to decide your fate in many scenarios - all because we are yet to work out something that better indicates how well you learned whatever you set out to learn.

TL;DR:

1. Stay away from DSF for the class size, stay away from AoA - it is not easy , amazing CV and NLP courses but are heavy because they take time.
2. Stay on track of for your assignments and project deadlines at all times , don’t leave it for later.
3. Get cracking on your resume and LeetCode before you arrive here.

Best of Luck \& Peace!