Computational Learning Theory¶
← Back to Domain-Specific Abstractions by Domain
3 domain-specific abstractions whose origin domain is Computational Learning Theory.
- Concept class — A family of Boolean-valued concepts or classifiers over a common instance domain, serving as the hypothesis universe in computational learning theory.
- Teaching dimension — The maximum, over concepts in a class, of the smallest labeled example set that uniquely identifies each concept to a learner when examples are chosen by a helpful teacher.
- Witness set — A set of input points whose labeled values distinguish one Boolean function or concept from every rival in a declared class.