“There is geometry in the humming of the strings, there is music in the spacing of the spheres.” – Pythagoras Introduction Recently I started exploring the world of error-correcting codes, and algebraic codes in particular. It’s a perfect example of mathematical hacking, where the Platonic realm of mathematical objects collides with the concrete world of signals, circuits, and CPUs to create something incredibly useful. In this post I’ll share what I’ve learned so far about algebraic codes and go into some detail about Reed-Solomon and BCH codes.

And everything else which can be attributed to body presupposes extension, and is only a mode of that which is extended. – Descartes, Principles of Philosophy Introduction gh is the official GitHub CLI, maintained in the public cli/cli repository. Like its predecessor hub1, gh wraps common git commands. But gh does much more, ultimately aiming to be a general GitHub interface for your terminal. You can for example run actions with gh workflow, publish gists with gh gist, or access your codespaces with gh cs.

Introduction The manifold hypothesis says that most real-world datasets lie approximately on a low-dimensional manifold, but by some Kantian twist of fate we rarely have direct access to this manifold1. As a result, most machine learning techniques utilize only local structure (e.g. the “loss + SGD” machine). In contrast, Topological and Geometric Data Analysis (TGDA) is a burgeoning field that studies the global structure of data. This is exciting not only because of potential domain applications – it also opens possibilities for porting a lot of powerful and beautiful math to the ML realm.