Skip to content
ML Academy

Introduced on day 56

Claim ladder

The ordered set of evidence a comparison claim needs - correctness, fairness, noise, relevance and limits - where the claim is only as strong as its lowest unsupported rung.

In detail

For a neural-versus-baseline comparison the rungs are: the network is correct (shape, gradient and tiny-batch checks), the comparison is fair (same split, features, metric and tuning budget, test opened once), the gap is larger than noise, it matters for a decision and holds on the key slices, and its limits are stated. Each rung needs an artefact; a rung without one is reported as not shown and the claim is cut back to match.