Statisticsmath

The mathematics of collecting, summarizing, and inferring from data — distributions, means, variance, hypothesis testing.

Evaluating compiler output (coverage, confidence distributions, misconception frequency) is statistical; understanding it prevents over-trusting single runs.

Where it appears

Descend

Prerequisites

Dependencies

Foundations

Historical evolution

From Fisher and Pearson to modern ML evaluation metrics.

Implementation details

SKCE reports coverage as a ratio and model-vs-deterministic node/edge counts in the manifest.

Sources

View in graph →