Nathan’s Notes

Ideology, beyond robustness and agency

Exploring the intricate web of word recognition in the brain, in relation to language models

The math behind why RNN-ifying SSMs works.

Why we should bet on open source humanoid robots.

Multi-armed bandits and a prolonged analogy

An accessible introduction to MRI Reconstruction and solving them with unrolled neural networks

An introduction to Mamba models: faster and better* than transformers

Discarding the cracked "Stanford CS" stereotype

A overview of diffusion models and flow matching models.

Thinking more about what's actually best.