Loading timeline…
20132010s
Tomáš Mikolov
Creator of word2vec, which turned words into vectors and made embeddings standard practice in NLP.
Organizations
GoogleFacebook AI ResearchCzech Institute of Informatics, Robotics and Cybernetics
Major Achievements
- •Led the word2vec work (2013), making it practical to learn word embeddings from very large corpora.
- •Showed that arithmetic on embeddings captures semantic relationships, the result that made vector semantics famous.
- •Earlier demonstrated recurrent neural network language models that outperformed the dominant n-gram approach.
Key Papers
- Efficient Estimation of Word Representations in Vector Space
Tomas Mikolov et al. · ICLR 2013 Workshop
- Distributed Representations of Words and Phrases and their Compositionality
Tomas Mikolov et al. · NeurIPS 2013