News for hackers. Stuff that matters.

A comment on Show HN: Basilica – word2vec for anything. Parent comment

Re:Show HN: Basilica – word2vec for anything by PeterisP on Wednesday October 31, 2018 @06:54PM (#18349876)
Do you plan to make these (many in the future) embeddings to refer to a single 'semantic vector space' or have each of them be separate?

I.e. do you plan to do the work to align the embeddings of different types of media so that the contents are somewhat similar and e.g. an audio recording of a snippet gets a similar embedding to the equivalent text and a somewhat similar embedding to a picture that's being described?

[ Parent | on HN ]
Re:Show HN: Basilica – word2vec for anything by mlucy on Wednesday October 31, 2018 @06:59PM (#18349912)
We aren't currently doing this.

In the future I think we'll try to embed into a single space on a best-effort basis, assuming we can find the engineering resources. It will be really hard for some data types, but for the big ones like text/image/audio it isn't that hard, and will probably be valuable to people.

[ Parent | on HN ]