On the class separability of contextual embeddings representations – or “The classifier does not matter when the (text) representation is so good!”
Cláudio Moisés Valiense de Andrade, Fabiano Belém, Washington Cunha, Celso França, Felipe Viegas, Leonardo Rocha, Marcos André Gonçalves
Topics & Concepts
Classifier (UML)Discriminative modelComputer scienceArtificial intelligenceNatural language processingRepresentation (politics)Separable spaceClass (philosophy)Pattern recognition (psychology)Machine learningMathematicsPolitical scienceMathematical analysisPoliticsLawTopic ModelingText and Document Classification TechnologiesNatural Language Processing Techniques