←back to thread

86 points alexop | 3 comments | | HN request time: 0.45s | source
Show context
frotaur ◴[] No.43328389[source]
As a physicist, I always found it funny that in ML people renamed 'the angle between vectors' into something as fancy-feeling as 'cosine similarity'.
replies(6): >>43328621 #>>43329354 #>>43329714 #>>43329823 #>>43330344 #>>43331512 #
1. Ultimatt ◴[] No.43330344[source]
I think the point is they're often doing it on things that aren't strictly vectors by definition, as in the physics sense, rather just some sack of stuff organised like they might be a vector and assuming its valid to compare angles because its just something numerically without much bias that does the job.
replies(2): >>43330733 #>>43330770 #
2. zorked ◴[] No.43330733[source]
It's stil a vector though.
3. frontfor ◴[] No.43330770[source]
I think what you meant to say is that they aren’t strictly geometric or “interpretable”. But they absolutely are vectors in the linear algebraic sense.