EasyDeepLearn

Word embeddings as unsupervised DR of text.

medium

Answer

  • Word2Vec (skip-gram, CBOW), GloVe: predict context words → learn dense word vectors capturing co-occurrence structure.
  • Word2Vec is implicit PMI matrix factorization (Levy & Goldberg 2014).
  • GloVe: explicit log-count factorization.
  • FastText: adds sub-word info for OOV.
  • Modern replacement: contextual embeddings (BERT, sentence-transformers) — the dominant text representation since 2018.
Check yourself — multiple choice
  • Random
  • Word2Vec / GloVe / FastText factorize co-occurrence matrix into dense word vectors; contextual embeddings (BERT) dominate since 2018
  • Same as t-SNE
  • Not DR

Word embeddings: dense DR of co-occurrence matrix; contextual embeddings modern.

#dimensionality-reduction#nlp#representation-learning

Practise Unsupervised Learning

214 interview questions in this topic.

Related questions