Maybe I love learning, or maybe I like going to class because I meet beautiful minds
A Mathematician’s Lament
This term, I’m studying Computer Vision with a prof that barely taught any computer vision. He spent 4 weeks teaching general image recognition technology, 4 weeks on high-dimensional math, and last week he sent us three files:
- The original research paper of the Black-Scholes model, a framework for pricing stock options and derivatives;
- The book “A Mathematician’s Lament“, where a mathematician gushes over his passion of math as an art, and how education reduces it to rote memorisation and pattern-recognition
- An extended discussion of “all models are wrong, but some are useful” in the context of financial modelling, with a 12-minute clip from the movie about the 2008 Global Financial Crisis, Margin Call, showing what happens when useful wrong models become unusefully wrong:
Nothing to do with computer vision. Barely anything to do with computers or machine learning. But there is some unifying theory, some equivalence between machine-learning models and financial-economics models, and some harmonic resonance (ha!) underneath the math and the science and the code.
That connection is beautiful. It’s like poetry, it rhymes; it’s like a pattern that repeats; it’s got themes and leitmotifs.
I always say that in the absence of LSD, learning computer science and algorithms is enough to blow my mind. This is my kind of high.
We can figure out the code ourselves, but this kind of learning and insight and revelations is why I really enjoy spending time with beautiful minds.