News for hackers. Stuff that matters.
Re:Show HN: Word2Bits – Quantized Word Vectors by isoprophlex on Tuesday March 20, 2018 @03:58AM (#16626693)
You can try encoding your input in shingles maybe, and feeding vectorized shingles (instead of a one hot encoded dictionary) into the usual CBOW or skipgram thing to train the embedding.

You'd need to invest plenty of effort into shingling and vectorizing properly to get useful results, though.

Re:Show HN: Word2Bits – Quantized Word Vectors by creichenbach on Tuesday March 20, 2018 @05:00AM (#16626930)
Ah, so we'd divide our word (fragment) into parts and treat the parts like words for usage with CBOW or skipgram?
Re:Show HN: Word2Bits – Quantized Word Vectors by isoprophlex on Tuesday March 20, 2018 @11:23AM (#16629333)
Uh yeah I have no idea if it'd perform well, but instead of having a sparse vector with the one-hot encoding of 'cargo', you enter a sparse vector with the 'car', 'arg' and 'rgo' dimensions set high.

Top of my head speculation, I never tried this...

Re:Show HN: Word2Bits – Quantized Word Vectors by creichenbach on Wednesday March 21, 2018 @08:54AM (#16637577)
Ah, that's actually not too far idea-wise from what we ended up doing: https://news.ycombinator.com/item?id=16637525