The Causal Learning Series: An Introduction
A guided, ground-up path through causality, the discipline behind every world model. Start with why correlation was never enough, and build toward models that reason about cause and effect.
Every world model rests on one idea: that a system can be described not by what tends to happen, but by why it happens. That idea is causality, and it is a discipline in its own right, with its own language, tools, and hard-won results. The Causal Learning Series is our attempt to teach it from the ground up.
Why a series
Causal reasoning is the part of machine intelligence that pattern-matching skips. A model trained only on correlations can tell you that two things move together; it cannot tell you what would happen if you intervened on one of them. That gap is exactly where enterprise decisions live, and closing it takes more than a single article.
So we built a sequence. Each chapter is short, self-contained, and concrete, moving from the intuition to the mechanics without assuming a background in statistics. Read in order, they add up to a working understanding of how a world model reasons about cause and effect.
Where to start
Chapter one asks the question the whole series turns on: why is causality necessary at all, when correlation has carried machine learning so far? It is the right place to begin.
Start the series: Why causality →
The Causal Learning Series is hosted in the Ergodic documentation. It updates as new chapters are published.
Run the decision before you commit to it.
Bring a decision. We'll discuss how to model it and what data a simulation would need.
Talk To An Expert