SESKStuff Everyone Should Know
SESK: Stuff Everyone Should Know · sesk.info/ai-decision-tree-white-box

Why are decision trees called a "white box" model?

You can trace a prediction through simple yes-or-no rules.

Start playing free

  1. Their rules always fit nonlinear data
  2. Their predictions ignore the training data
  3. Each prediction follows readable if/then logic ✔

Why: A decision tree is a chain of conditions, like "is age over 30?", ending in a leaf. You can read the path for any prediction, unlike a neural network's millions of weights.

Huh, didn't know that: The catch: trees grown too deep overfit easily, especially with many features.

The dad joke
My decision tree asked me twenty yes or no questions and decided I wanted pizza. It was right.
Like this card?
Play the game

References

Answer: Each prediction follows readable if/then logic. A decision tree is a chain of conditions, like "is age over 30?", ending in a leaf. You can read the path for any prediction, unlike a neural network's millions of weights.

  1. scikit-learn User Guide: Decision Trees scikit-learn.org/stable/modules/tree.html
    Uses a white box model. ... the explanation for the condition is easily explained by boolean logic.
    Checked 2026-10-10.

Think this answer is wrong? Tap "Challenge this answer" on the card in the game and tell us why.

Test yourself in the game

Artificial IntelligenceHow is a random forest built?Artificial IntelligenceHow does NIST distinguish explainability from interpretability?Artificial IntelligenceHow do AI, machine learning and deep learning fit together?

That's stuff everyone should know.