Occam learning (original) (raw)
In computational learning theory, Occam learning is a model of algorithmic learning where the objective of the learner is to output a succinct representation of received training data. This is closely related to probably approximately correct (PAC) learning, where the learner is evaluated on its predictive power of a test set. Occam learnability implies PAC learning, and for a wide variety of concept classes, the converse is also true: PAC learnability implies Occam learnability.
| Property | Value |
|---|---|
| dbo:abstract | In computational learning theory, Occam learning is a model of algorithmic learning where the objective of the learner is to output a succinct representation of received training data. This is closely related to probably approximately correct (PAC) learning, where the learner is evaluated on its predictive power of a test set. Occam learnability implies PAC learning, and for a wide variety of concept classes, the converse is also true: PAC learnability implies Occam learnability. (en) Оккамово обучение в теории вычислительного обучения является моделью , где целью обучения является получение сжатого представления имеющихся тренировочных данных. Метод тесно связан с почти корректным обучением (ПК обучение, англ. Probably Approximately Correct learning, PAC learning), где учитель оценивает прогнозирующую способность тестового набора. Оккамова обучаемость влечёт ПК обучение и для широкого класса понятий обратное тоже верно — ПК обучаемость влечёт оккамову обучаемость. (ru) |
| dbo:wikiPageID | 44577560 (xsd:integer) |
| dbo:wikiPageLength | 10840 (xsd:nonNegativeInteger) |
| dbo:wikiPageRevisionID | 1085332514 (xsd:integer) |
| dbo:wikiPageWikiLink | dbr:Probably_approximately_correct_learning dbr:Deterministic_finite_automaton dbc:Computational_learning_theory dbr:Concept_class dbr:Computational_learning_theory dbc:Machine_learning dbc:Theoretical_computer_science dbr:Occam’s_Razor dbr:Structural_Risk_Minimization |
| dbp:wikiPageUsesTemplate | dbt:Reflist dbt:Short_description dbt:Machine_learning |
| dct:subject | dbc:Computational_learning_theory dbc:Machine_learning dbc:Theoretical_computer_science |
| gold:hypernym | dbr:Model |
| rdf:type | dbo:Person |
| rdfs:comment | In computational learning theory, Occam learning is a model of algorithmic learning where the objective of the learner is to output a succinct representation of received training data. This is closely related to probably approximately correct (PAC) learning, where the learner is evaluated on its predictive power of a test set. Occam learnability implies PAC learning, and for a wide variety of concept classes, the converse is also true: PAC learnability implies Occam learnability. (en) Оккамово обучение в теории вычислительного обучения является моделью , где целью обучения является получение сжатого представления имеющихся тренировочных данных. Метод тесно связан с почти корректным обучением (ПК обучение, англ. Probably Approximately Correct learning, PAC learning), где учитель оценивает прогнозирующую способность тестового набора. Оккамова обучаемость влечёт ПК обучение и для широкого класса понятий обратное тоже верно — ПК обучаемость влечёт оккамову обучаемость. (ru) |
| rdfs:label | Occam learning (en) Оккамово обучение (ru) |
| owl:sameAs | freebase:Occam learning wikidata:Occam learning dbpedia-ru:Occam learning https://global.dbpedia.org/id/2MPQo |
| prov:wasDerivedFrom | wikipedia-en:Occam_learning?oldid=1085332514&ns=0 |
| foaf:isPrimaryTopicOf | wikipedia-en:Occam_learning |
| is dbo:wikiPageRedirects of | dbr:Occam_Learning |
| is dbo:wikiPageWikiLink of | dbr:Probably_approximately_correct_learning dbr:Ockham dbr:Occam_Learning dbr:Computational_learning_theory dbr:Outline_of_machine_learning |
| is foaf:primaryTopic of | wikipedia-en:Occam_learning |