VL12 NL1




What is ๐Ÿง‘ BERT?#card

= a ๐Ÿง  Machine Learning technique for โš™ Natural Language Processing, which uses sparse, contextu^1642441138524 al embeddings.

Define ๐Ÿ‘€ Contextual embedding#card

= ใ€ฐ๏ธ Word embedding that takes the context of a wor^1642441138535 d into account

Define ๐Ÿคทโ€โ™‚๏ธ Non-Contextual embedding#card

= ใ€ฐ๏ธ Word embedding that does not take the context of a wor^1642441138546 d into account

Whatโ€™s the ๐Ÿ‘” Chomsky normal form?#card

= a ๐Ÿ“– Formal grammar, which fulfills the follo^1642441138555 wing criteria:

no empty right-hand side every right-hand side has either 2x non-terminals or 1x terminal

Whatโ€™s a ๐Ÿ— Formal generative grammar?#card

= a fully explicit ๐Ÿ“– ^1642441138563 Formal grammar

Whatโ€™s the formal definition of a ๐Ÿ— Formal generative grammar?#card

  • : finite set of non-terminal symbols
  • : finite set of terminal symbols disjoint from
  • : start symbol,
  • : finite set of rules of type
    • conaining min 1 non-ter^1642441138570 minal symbol

List 3/4 Phenomena#card

  • Subcategorization
    • verbs can take certain complements
    • but can also have number of optional modifiers
      • e.g. โ€œI sleep hereโ€, โ€œI hardly ever sleep hereโ€
  • Agreement
    • main verb agrees with person and number
      • e.g. โ€œWe drive a carโ€, โ€œHe drives a carโ€
  • Movement
    • certain constituents can be moved around
    • e.g. โ€œHe drove backโ€, โ€œBack he droveโ€
  • Coordination
    • phrases can be conjoined by means of and and or
    • โ€œJohn and Ni^1642441138578 ck drove homeโ€