← All articles

Research notes: from equations to experiments

A place for clear derivations, small reproducible experiments, and implementation notes on generative modelling, reinforcement learning, and optimization.

This blog brings together notes on generative modelling, reinforcement learning, and optimization. The aim is to make the connection between a mathematical formulation and its implementation explicit.

A useful explanation should answer three questions:

  1. What is the claim? State the result and the assumptions under which it holds.
  2. Why does it hold? Work through the argument, including the steps that are easy to skip.
  3. What changes in code? Use a small experiment to check the reasoning and identify where approximations enter.

Topics will include continuous-time dynamics, transport and density evolution, efficient generative inference, and learning objectives. Some posts will be short explanations of a single equation; others will connect a derivation to an implementation.

Research projects and publications are collected on the research page and papers page. This space is for the explanations and experiments around them.

You can follow new articles through the RSS feed.