I recently decided to leave a graduate program at Johns Hopkins University to pursue my interests in industry. This was not a decision I took lightly. My biggest reason for heading down this new road involved a growing skepticism about how well the training I was receiving would prepare me for the kinds of jobs I’d eventually like to pursue, but several important factors influenced my reasoning about this very big change.
Finding my academic identity
My driving passion is my interest in the mechanics of attention, learning and decision making in both biological and artificial systems. While it might seem obvious to reason that our understanding of the human brain might do much to inform our progress toward an artificial general intelligence, in grad school I encountered many people who scoffed at the idea that neuroscience might have any more wisdom left to impart to computer scientists. And it’s true – I’ve come to understand that brains and machines are more different from each other than I had imagined. But that doesn’t mean we should stop looking to the brain for inspiration for A.I., or vice versa.
I want to be a part of the realization of the first artificial general intelligence, even if only in some small way, some lesser-known paper published in some open-access journal maybe. But I think I can do at least that and more. I think I’m finally qualified to begin developing quantitative models of networks of attention and decision making, and I’d like to apply what I learn to practical commercial or industrial products.
Toward the end of my first year at Hopkins, it began to become clear to me that the aim of my career trajectory was a bit off target. I ended up taking a course called Quantitative Methods for Scientific Data Analysis, taught by my academic advisor, and it woke a nascent but powerful fascination for the marriage of statistical learning and computer programming. I found it immensely satisfying to use computers to simulate and inspect data, vast amounts of data too, and I didn’t even have to go through the usual scientific drudgery of collecting it through long experiments! Even better, I loved being able to test theories immediately and iteratively, without any need to wait for data collection or red tape.
At the same time, I was learning about scientists like David Marr, a psychologist/neuroscientist whose lasting impact on artificial intelligence is still felt today; I was reading papers published in computer science journals about all kinds of exciting advances in the field, including everything from new models of vision/language recognition to renewed popular excitement in models of statistical learning like neural networks; I was learning that I too wanted to pursue these kinds of big ideas at the intersection of computer science, philosophy, math and psychology.
In my grad program at Hopkins, I was certainly picking up some incredible skills, but I became skeptical about whether or not they were specific enough to my interests. It seemed like I was spending too much time doing things like surgery and experiments on animals, which I loathed, and I saw a widening gulf between what I was specializing in and what I wanted to specialize in. I knew that if I didn’t correct my course now, at the near-beginning of my journey, then I would regret it later on.
But something else happened in my life toward the end of my first year of grad study: I got engaged! Exciting as this personal development was, it presented another problem for my career. My beautiful fiancee has worked very hard to build her career in California, but since I expected to be stuck in Baltimore for five more years, our engagement meant that she would have to drop her achievements and rebuild anew on the east coast with me. At first, that was just what we planned to do, but slowly dawned on me that this plan really didn’t make sense. Why should the partner who is more established and invested in her career drop everything for the partner who is just beginning?
This realization was the nail in the coffin for my studies at Johns Hopkins. The combination of the doubt about my path of study and the needs of my personal life made it clear to me that it was time for me to make a change. Moving to California presented the opportunity to support my fiancee while also simultaneously refining and revitalizing my career plans. I’ll write in detail about my current plans and intellectual objectives in an upcoming blog post, but for now, for the record, I am extremely excited about both my near- and long-term goals and opportunities, and I look forward to chronicling more of my journey here soon.