Fully Character-Level Neural Machine Translation without Explicit Segmentation · HackerPapers discussions

Fully Character-Level Neural Machine Translation without Explicit Segmentation — HN submission

Mon, 02 Jan 2017 21:51:58 GMT

Fully Character-Level Neural Machine Translation Without Explicit Segmentation

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

Fully Character-Level Neural Machine Translation without Explicit Segmentation — mentioned by stared

Mon, 02 Jan 2017 20:49:23 GMT

Seconding it with: https://arxiv.org/abs/1610.03017 Fully Character-Level Neural Machine Translation without Explicit Segmentation. In general, RNNs are especially fit for discrete sequences. And their continuous representation is actually an advantage (so they can see that two words are similar or analogous). BTW: see my draft on word2vec: http://p.migdal.pl/2016/12/30/why-do-word2vec-analogies-work...

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

Fully Character-Level Neural Machine Translation without Explicit Segmentation — HN submission

Thu, 13 Oct 2016 12:30:27 GMT

Fully Character-Level Neural Machine Translation Without Explicit Segmentation

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

Fully Character-Level Neural Machine Translation without Explicit Segmentation — HN submission

Tue, 11 Oct 2016 23:35:13 GMT

Fully Character-Level Neural Machine Translation Without Explicit Segmentation

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