About

You’ve probably heard of the infinite monkey theorem: Given endless time, a monkey typing randomly could produce all of Shakespeare’s works. But what happens when we give an ape a computer and a finite amount of time? You are about to find out. (Spoiler: I am no Shakespeare.)

I’m Berk Idem. By day I do machine learning and economics.¹ I spend the rest of my life in rabbit holes, some of which have grown to Herculean proportions. This blog is where I think those things through in writing, mostly for my own sake, and partly to find the handful of other people who are interested in the same kinds of problems.

My interests are broad, but they generally fall into the following categories. The categories are, as is often the case, fairly arbitrary. I tend to think of the social sciences as one extended subject and applied mathematics as a densely connected cluster, so I am usually most interested in the borders and overlaps between fields.

• Mathematics, Logic & Computation: Mathematics, logic, and the theory of computation. Lately a long reading group on the foundations of logic and mathematics has taken over, running from Dedekind, Cantor, and Frege up through Russell, Peano, Zermelo, and beyond.
• Philosophy & History: Mostly analytic philosophy, lately the philosophy of language (yet another reading group) and the philosophy of mathematics, plus whatever historical context pulls me in. Somewhere along the way I started learning ancient Greek.²
• Science, Applied Math & Technology: Game theory, evolution, linguistics, machine learning, operations research, intelligence (of any kind), causal inference… I also tinker with small single-board computers and write little apps for myself. I keep a foot³ in applied math and work through many of its subfields: dynamical systems, information theory, transport theory, etc.

For every post I publish, I probably write four or five more that never escape draft status. Sometimes I am not sure anyone else would care; sometimes I get the dopamine hit from writing them and stop before publishing. I am trying to change this. LLMs have made editing cheaper, but thinking remains stubbornly expensive, so we’ll see how it goes.

I also have some long-term writing projects but don’t ask me about any deadlines: a novel (historical science fiction-y), some notes on reinforcement learning for economists (or for any dynamic programming-knowers, I suppose), and another set of notes on machine learning for economists with an amazing co-author (who is human).

More broadly, I want to make the connections among subdisciplines easier to see. Over the years (more like decades at this point…), I have written many notes for myself to make those connections more explicit. Some of them may eventually escape draft status. They might be especially useful for economists, but really for anyone with a decent applied math background and, as they say in the prefaces, mathematical maturity.

Having said so much about what to expect, let me also tell you a little about what I don’t expect to write: This won’t be a place for “10 best bullet journal ideas to organize your life”. I still don’t know what bullets have to do with journals.⁴ I’ll also try to spare you my fermentation and baking experiments, but I can’t promise anything. You might get the occasional book-nerding post.⁵ And you may also see some footnotes that compete with the main text in length. The theme makes it easy to ignore the footnotes (or “sidenotes” if you are not on a vertical screen). Just wanted to give you a heads-up about what you’re getting into.

If you want to say something, the contact page is over there.