3 ms·

One thing with it is the distance from getting a computer to compose anything to getting it to compose close to human level could be 90% of the distance to gett

One thing with it is the distance from getting a computer to compose anything to getting it to compose close to human level could be 90% of the distance to getting it to compose at a super human level. Even several months before Alpha Go beat Lee Sedol there were people saying the AI could be good but never great. Now everyone admits it's super human. I agree that getting really texture and subtlety in the work is really hard. However a really good system has the possibility to be better than any human has ever been, and IMO that's super exciting.

abeppu 7y ago
What does "super human" mean for a cultural artifact like music? With a game, "super human" can mean "beats humans". For supervised learning, where there's some label other than human opinion, "super human" can mean "more accurate than a human labeler". But for music composition, you need a difference source of information (listener ratings? sales/streaming plays?) to determine what "better" means, which would be a meaningful structural change rather than just an extension of this technique. So long as the training set only tells it what human-composed compositions look like (MIDI files of existing music), a system has no signal to find "super human" territory.
- drchewbacca 7y ago
  You're right measuring the skill of a composer is hard, however as you say listener ratings and sales/streams would be a pretty good metric. Even for songs in the top 100 charts how high can an AI get? If all the top 10 songs were written by an AI that would be pretty convincing it was super human I think. On the issue of generalising alpha go learned go by watching human played games and by playing against itself and is now better than any human. So it's not true that a system is limited only to the skill of the examples it's shown, it's possible to design something which can surpass the examples it learned from.

abeppu 7y ago

What does "super human" mean for a cultural artifact like music? With a game, "super human" can mean "beats humans". For supervised learning, where there's some label other than human opinion, "super human" can mean "more accurate than a human labeler". But for music composition, you need a difference source of information (listener ratings? sales/streaming plays?) to determine what "better" means, which would be a meaningful structural change rather than just an extension of this technique. So long as the training set only tells it what human-composed compositions look like (MIDI files of existing music), a system has no signal to find "super human" territory.

drchewbacca 7y ago
You're right measuring the skill of a composer is hard, however as you say listener ratings and sales/streams would be a pretty good metric. Even for songs in the top 100 charts how high can an AI get? If all the top 10 songs were written by an AI that would be pretty convincing it was super human I think. On the issue of generalising alpha go learned go by watching human played games and by playing against itself and is now better than any human. So it's not true that a system is limited only to the skill of the examples it's shown, it's possible to design something which can surpass the examples it learned from.