Enriching Word Vectors with Subword Information · HackerPapers discussions

Enriching Word Vectors with Subword Information — mentioned by physicsyogi

Tue, 20 Mar 2018 14:00:12 GMT

Enriching Word Vectors with Subword Information: https://arxiv.org/abs/1607.04606

Paper and discussion history. A mention is not an endorsement.

Enriching Word Vectors with Subword Information — HN submission

Tue, 04 Jul 2017 21:59:30 GMT

Enriching Word Vectors with Subword Information [pdf]

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Enriching Word Vectors with Subword Information — mentioned by minimaxir

Thu, 25 Aug 2016 20:22:52 GMT

This was posted in 2015, but now that fasttext (https://github.com/facebookresearch/fastText), just released by Facebook and can scale to Instagram-sized datasets, can create word vectors better than word2vec which account for context (https://arxiv.org/pdf/1607.04606v1.pdf), this type of analysis will only improve in the future.

Paper and discussion history. A mention is not an endorsement.

Enriching Word Vectors with Subword Information — mentioned by kough

Thu, 04 Aug 2016 20:29:27 GMT

Links to the relevant papers: Bag of Tricks for Efficient Text Classification: https://arxiv.org/abs/1607.01759v2 Enriching Word Vectors with Subword Information: https://arxiv.org/abs/1607.04606 Both fantastic papers. For those who aren't aware, Mikolov also helped create word2vec. One curious thing: this seems to use heirarchal softmax instead of the "negative sampling" described in their earlier paper http://arxiv.org/abs/1310.4546, despite that paper reporting that "negative sampling" is more computationally efficient and of similar quality. Anyone know why that might be?

Paper and discussion history. A mention is not an endorsement.