Version Space Learning

Fact

A set of candidate rules that still match the labeled examples.

In Plain Words

Version Space Learning is like Guess Who? for AI models. Each labeled example flips down more little faces.

You meet it in supervised learning and active learning. It shows how examples shrink the model choices.

Related Concepts

• Supervised Learning
• Inductive Bias
• Active Learning
• Probably Approximately Correct Learning

Supervised Learning
Version Space Learning uses labeled examples to drop rules that do not fit.

Inductive Bias
Inductive Bias sets the starting pool of possible rules.

Active Learning
Active Learning can pick examples that shrink the version space fastest.

PAC
PAC also asks how many examples can rule out wrong rules.