Even as I stepped foot into George Mason University, wide-eyed about the prospect of coming across Digital Humanities research, I fell harder in love with my department itself. Here I was, in the middle of the department orientation with barely any students but oh, the researchers there – Paleontologists, Geologists, Quantum Physicians, Operations Researchers, you name it. I would come to know these people for their research from the weekly Center for Social Complexity’s Computational Social Sciences Research Colloquium Seminars. I would go on to take a few of their classes. It was overwhelming to actually bear witness to people who were more fiction than fact from oceans away, people whose passions I related to more closely than any Computer Science department. They knew the potential for multi-disciplinary studies. I admit being attracted to the vision of studying multiple topics, whatever caught my fancy, however I stumbled across it, just as the ancients of Platonic Academy, or better yet, Nalanda University or Takshashila University. And yes, 22 year old Jaj, I did.

References to Hitchhiker’s Guide to the Galaxy, Doctor Who notwithstanding, CSS 600 was my favourite class from my first semester (fall 2019) at GMU, and it was because of the instructor. Apart from being a knowledgeable and pleasant instructor, Dr. Andrew Crooks is encouraging towards his students. He takes the time to listen, understand, and advise students in his class and outside for seminars, conferences, etc. He agreed to recommend me for summer internships, fellowships, and job prospects and I am extremely grateful to him for that. He helped me consider different domains – Where I was previously fixated on Digital Humanities, he helped me learn about research centers, think tanks, and academia in Computational Social Sciences.

For the Spring 2020 semester I picked the sequel to CSS 600 – CSS 610 and I simply would not stop harping to friends, about just what a stalwart in the field I was going to be taught by. Dr. Robert Axtell’s classes had me in thrall. The first few weeks turned out to be an extremely sped-up version of what amounts to around 7 or so undergrad Economics courses. Not only was I trying to code consistently in Python and NetLogo but I was trying to understand all these terms Dr. Axtell was throwing around: macroeconomics, suppy, demand, Tragedy of Commons, game theory, Nash equilibria, all churned in my heavy head, and calcified into a greater wonder about the world around me.

Math and me. We have an uneasy relationship. You could say it has become quite unhealthy. I was never quite natural at it. I spent hours, waking up at 4am, in my third grade to learn the multiplication tables. I practiced each problem from the textbook a couple of times before I felt confident about taking up an exam. When I took up Pre-University electives as Science, PCMCs (Physics, Chemistry, Mathematics, and Computer Science) I knew Calculus was not going to be easy. I filled up notebooks with the trigonometric functions. When the entrance exams to engineering college placement came up, I had to focus on speeding up my calculations and understanding of the questions themselves. I can confidently say I put my blood, sweat, and tears into math this time. And then, as a Computer Science and Engineering student, I had two years of Engineering math, Discrete Structures math, Logic Design math, Graph Theory and Combinatorics, and Operation Research math… where I further deemed myself not smart enough for math. 18 year old me could bust out a 1000 word short story and win Creative Writing competitions for it but to enjoy math was to twist an ankle but become an overnight Muay Thai sensation overnight; incomprehensible. I wish I could have told younger me that the techniques, methods, and form of Mathematics I was trying to grasp was not beyond me; it was just not the way I should have learnt it. It was not the way Mathematics can be learned. I always perceived the beauty of mathematics in a distant unseen way but could only really know what I was missing after I started reading nonfiction science books like Carl Sagan’s Cosmos and Broca’s Brain or watched films like Mindwalk. They were influential in helping me consider that Maths wasn’t all that it was made out to be in dry dreary characterless context-less textbooks. Mathematics has a cadence to it, and has to be learned like a language. After all, programming languages are a combination of the written(typed) words and equations for just this reason.

Now working out mathematics problems became a matter of philosophy. Nietzsche-ian – Viktor Frankl-ian to be exact. “Know the why to suffer the how.” Trying to understand why the area of a circle is πr^{2 }became important suddenly. Animated mathematical concepts made a world of a difference in how I perceived the visual logic of such equations. YouTube channels like https://www.3blue1brown.com/ and https://thecrashcourse.com/courses helped. By the time I was giving my GRE my mindset regarding maths had changed. That is not to say that, I magically, suddenly, fantastically, became excellent at problem solving. The title to this post is not “I used to fail in math. Here’s how I aced the GMAT/CAT!” I did nothing of that sort. My GRE score was a so-so 150 in the Quant section. Just about within my goal range.

At each stage of my educational career, I swore off Mathematics once and for all. *After this test / exam / class / year, I am never touching academic maths as a subject ever again*,I would tell myself. Each time, it wouldn’t be long before I was swearing off maths yet again.

As a Data Scientist, I figured I could pick a field and profession within, with minimal maths in it. Hello Digital Humanities. I thought I could just rough it a few times. I would have to work with maths and maybe coax / cajole it to work with me. Of course, when I got into the Computational Data Sciences masters program, I fully expected to encounter the roughing up stage initially. Even so, Numerical Methods snuck up on me with its intensity. Not only was I to become comfortable enough to play around with the fundamental mathematical theories for machine learning concepts, but also adapt them into a computational form with… Python. I had dug myself into the deepest hole yet. The pinprick of sunshine in this tunnel was all the help I asked for. Vishwanath, an absolute smarty from a group of online fandom geekery, Fandoms United India, has my eternal gratitude.

Every week, I hunted down every resource possible to better understand Least Squares Method, Taylor’s series, lower–upper decomposition or factorization, linear algebra, Principal component analysis, and ordinary then partial differential equations. I had bitten off more than I could chew, I knew. I went to the formidably brilliant professor repeatedly during all her office hours. I contacted the department and found former students, PhD students who, unbelievably kindly helped me understand the problems. I found a student from the Physics department taking up an incredibly similar course as Computational Physics, but in MATLAB, and became study buddies with her; how grateful I am for her. I whined my way through the semester with my fellow classmates and friends. The midterms and finals whizzed by, and I received a B. Which was dingdangkachingchang brilliant, according to me. Masochistic as it may seem, I became determined to take at least one math-based course every semester. It could have been Time-series analysis or Bayesian inference. Confused between the wide array of courses available to me, I attended a few preview lectures from different departments before I fixated on a Stat 515 course, Applied Statistics & Visualization for Analytics. Visualization with R? Check. Basics of Machine Learning algorithms visualized? Check! Mathematical formulae? Yep. Multiple birds with course and all that.

I suppose it has been more interesting than I anticipated. The pleasure of knowing how to perform a regression on a dataset, exceeds all.