Cuestionario vinculado a una guía · Medio Nivel

Prueba de aprendizaje autosupervisada

Revise los objetivos derivados de datos, el enmascaramiento BERT, los pares SimCLR, las necesidades de anotación, la adaptación, las tareas del siguiente token, los cabezales de proyección y los objetivos de tokens corruptos.

Rutas de guía relacionadasAprendizaje autosupervisado
Pregunta 1 de 8

Where does the supervision signal in the described self-supervised examples come from?

Vista previa de cada pregunta8 preguntas

Las respuestas y explicaciones aparecen a medida que realiza el cuestionario.

  1. Where does the supervision signal in the described self-supervised examples come from?
    • A requirement for no training objective
    • Only new human semantic labels for every pretraining example
    • A rule that examples contain no useful information
    • The data and defined transformations of it
  2. Which input positions contribute prediction targets to the original BERT masked-language-model loss?
    • The positions selected for the MLM prediction task
    • Every position in the entire input regardless of selection
    • Only the first position of each new sentence
    • Only positions that appear after the final sentence boundary
  3. En el aprendizaje contrastivo como SimCLR, un "par positivo" es:
    • Dos imágenes completamente ajenas
    • Un par de etiquetas escritas por humanos.
    • Dos vistas aumentadas de la misma imagen que deberían estar juntas en el espacio de incrustación
    • Dos copias del conjunto de datos.
  4. What annotation requirement can self-supervised pretraining reduce in the described examples?
    • The need for any usable input data
    • Separate task-specific semantic labels for every pretraining example
    • All computational cost by definition
    • The need to assess downstream performance
  5. How can a representation learned through self-supervision be used for a later task?
    • By evaluating a task head or adapting the representation through fine-tuning
    • By assuming every target task requires no labels or validation
    • By discarding all representations before making predictions
    • By treating the pretraining loss as proof of every application's accuracy
  6. ¿En qué se diferencia la tarea autosupervisada de un modelo de estilo GPT de la de BERT?
    • No utiliza ningún texto.
    • Requiere etiquetas manuales para cada palabra.
    • Sólo predice la primera palabra de un documento.
    • Predice el siguiente token en lugar de completar tokens enmascarados.
  7. In the original SimCLR framework, which component is discarded for the reported downstream representation use?
    • The encoder representation itself
    • The contrastive projection head
    • All training inputs before pretraining begins
    • The positive-pair source-image relationship
  8. When original BERT pretraining replaces a selected input position with a random token, what remains its loss target?
    • The newly inserted random token
    • The next sentence's first token in every case
    • The original token before replacement
    • No target, because every random replacement is omitted from training