Chance sometimes favors the unprepared mind too

Artificial IntelligenceComplexity & Simulation

🏄‍♂️

This guy does not look open-minded.

In fact, when this picture was taken, I (yes, it’s me, I haven’t changed a bit) had just started my first semester at École Polytechnique Polytechnique, a French science and engineering college, and I was already convinced that learning more science was probably a waste of time as I was dead set on applying to the French School of Civil Service (ENA). I was studying economics, history, literature and law on evenings and weekends to prepare for the competitive exam. But in the spring and summer of 1989, two serendipitous events happened in sequence that would change everything.

The first was an aesthetic shock: around April, I found incredible beauty and elegance in the most unlikely place, Introduction to Probabilities 101 by the extraordinary Jacques Neveu (1932-2016). That I would find such a topic enticing was not part of the plan. By June, I was beginning to question my commitment to prepare for ENA. And then, in July, another shock.

I was at UC San Diego for the summer programs, having picked (before April) US Constitutional Law as my summer course. Vivek Badrinath had orchestrated quite a coup, getting our college to cover the cost of the program. So I went but failed to find the classroom on Day 1, wandered around campus, ended up in the wrong place (corporate law), ran away and took refuge at UCSD’s bookstore. Browsing random sections, wondering whether my life was still on track, I stumbled upon a thick double-volume set entitled Parallel Distributed Processing (Explorations in the Microstructure of Cognition), by David Rumelhart and James McClelland. Why, I am not sure. But what I know is that I didn’t attend any classes at all for the rest of the program. I read (I tried to read) the two books, understanding about 5% of their contents, but feeling like I had found meaning in my life: connectionist models and neural networks, a perfect combination of cognition and applied math.

Upon my return, I gave up my plan to apply for ENA and embraced neural networks as the love of my professional life. For a while.