Goodbye Boston University

I think I wrote somewhere that his blog would be mostly technical with little other riffraff, but I don’t graduate from school all that often so I figured it would be OK to indulge in a post about other stuff. I have more legitimate topics I want to write about, I just haven’t found the time for it yet with school wrapping up.

The purpose of this is to put down some of my thoughts on Boston University as a school and reflect on the PhD application process. So I’ll do those in order!

Overall I enjoyed my time at BU and thought I got a lot out of it. I originally decided to go here because 1] it was in the range of places I could get into and 2] because it had a relatively strong Engineering college as well as Philosophy department, which was important to me. I think both of those things were mostly well-founded. Random thoughts:

• The Philosophy Department isn’t huge, but it in my opinion is very tight-knit and generally strong. I’d like to highlight Professors Floyd, Hopp, and Gasser-Wingate as exceptionally strong lecturers in particular. I still think about Wittgenstein regularly because of the class about him I took from Floyd. The Undergraduate Philosophy Association (the philosophy club) and Arche (the undergraduate student journal) are now both great institutions and I met many exceptional fellow students through them.

• The people in Machine Learning, especially on the control-side of things (robotics, reinforcement learning, etc.) is kind of all over the place because of strange organization structure, which makes things a little odd. BU is one university split up by college, and at least three colleges have people in the ML for control areas: the College of Engineering (through Computer Engineering, Electrical Engineering, Mechanical Engineering, and some Systems Engineering), the College of Arts and Sciences (through Computer Science), and the new Data Science college. The result is that through if you are interested in studying this stuff at BU, it can be a little hard to get them to fill the right course requirements for your given major because the relevant classes might be offered in a different college, where you would need to petition for the classes to count for a given elective credit. A small example of this ridiculous bureaucratic arrangement is that BU currently has (at least) three people working on the theory behind Markovian decision-making - Professor Alex Olshevsky, who is in ECE but has affiliations with Systems and CS, Professor Xuezhou Zhang, who is in DS but who advises an ECE and SE PhD students, and Professor Andrea Lincoln who is in CS (technically her work is more on complexity theory than learning algorithms, but she does have work on graph theory that’s pretty similar to the Markovian setting. So three professors, all working on essentially the same quite niche thing, in three different colleges, including a total of 5 departments, all at the same school. This seems a little silly to me. But it worked out for me in the end through a little petitioning (thank you Dr. Nazer) so c’est la vie.

• The Student Activities Organization (SAO), the people who manage undergraduate clubs, are useless at best and actively vile on average (I will spare you a superlative for at worst). The most struggle the UPA and Arche have recently had all came from SAO interactions. You can’t avoid interacting with them entirely as a student organization, so I will just say to avoid them as much as humanely possible.

• Regarding getting two degrees across two different colleges (the ‘dual degree’ program) - possible but somewhat difficult. The CAS language requirement is a particular point of annoyance (I have to take more Mandarin during the summer at SDSU because of it) but if you speak a non-English language natively and are doing a degree with not that many total required classes (Philosophy has an absolutely shockingly low number at just 9(!)) it is quite doable. Doing anything in ENG and an outside degree is hard though, just because ENG has such a tightly-packed course schedule already. But I don’t regret doing it for a minute. I will say the logistical hassle of writing the application, filling out the right paperwork, getting advisor signatures, etc. was more of a pain than it needed to be.

• Regarding the myth that if you step on the seal by Marsh Chapel you won’t graduate in four years - I never stepped on it, and am still ending up graduating a couple months late (although still technically within 4 years). Make what you will of that.

Misc. thoughts on CS PhD applications. Take with a grain of salt, I’m a single sample on only one side of the process etc. etc. :

• Applying wide seems smart. I was a strong student with mediocre research experience that was outside my desired area, and I was applying into a saturated field (RL), so I knew it would be rough. Excluding one safety option, I applied to 13 schools that were not in the top 4 CS schools, but roughly followed the schools after that with some give and take for their specific strength in RL. I didn’t apply to schools in the UK for funding reasons, or schools in Asia because while there are very strong schools there, its unusual for Americans who are not East Asian (or Americans at all, really) to go there and I doubt I would’ve had a great time. So 13 schools, widely regarded in RL, but not top 4, across the US, Canada, and Switzerland. I got into 2, so not a great yield, and the one thing that saved me was applying to a lot of places.

• Admissions seem quite random. A rough glance at my application sheet post-facto shows out of the 2 schools I got into, there were comfortably 2 and maybe more schools I would consider less prestigious/whatever than the 2 I did. There are maybe another couple one could argue are below the school I ended up committing to. I was certainly surprised at the final outcome. I think the first thing about applying wide helps alleviate some of this randomness.

• Going into the process I thought applying to schools outside the United States could help give me options, because I perceived some of these schools as being as strong academically but having less aggressive admissions competition as the top ~15 or so schools in the States. Domestically (I am a US citizen) I went 1/9, and internationally 1/4. So my impulse was good in the sense that it got me into where I committed to, bad in the sense I didn’t see great net outcomes from either.

• But end of the day, I am extremely happy about where I’m going and am still shocked I managed to trick the admissions committee at a decent school. Their loss!

To wrap up I just wanted to acknowledge some of the people who made my BU experience great. Research collaborators/advisors: Dr. Sharifzadeh, Dr. Coskun, Dr. Haldar, Fatih Acun. Professors: Dr.s Olshevksy, Ohn-Bar, and the Philosophy people I’ve mentioned above.

So long and thanks for all the fish! I will catch you all from the Six.