Interesting point of view. Will neural networks (or other state-of-the-art methods) end up refuting Occam’s razor, or will they turn out to be “simple” after all?https://twitter.com/pmddomingos/status/1392321198566961155 …
-
I was thinking the same thing as I Tweeted it, but maybe there’s some global sense in which all these algorithms are “simple” that we haven’t discovered yet. Or is that too much to hope for?
There's no universal notion of simplicity. Better to think in terms of hypothesis space size (or prior entropy), but even that is only an upper bound on the amount of search you do. See: https://homes.cs.washington.edu/~pedrod/papers/ijcai99.pdf …
-
-
And yet we all probably still favor simpler theories when it comes to physics, biology, our lives, etc., so how serious are you about Occam’s razor really being dead?
-
Dead serious :) Seriously, though, there's a different reason to prefer simplicity: lower cognitive/computational load. But it's a tradeoff between it and accuracy. See: https://homes.cs.washington.edu/~pedrod/papers/dmkd99.pdf …
- Näytä vastaukset
Uusi keskustelu -
Lataaminen näyttää kestävän hetken.
Twitter saattaa olla ruuhkautunut tai ongelma on muuten hetkellinen. Yritä uudelleen tai käy Twitterin tilasivulla saadaksesi lisätietoja.