here’s a conjecture: biological learning is never unsupervised. Rather, its goal is to learn representations that are useful for future behavior. @AstroKatie @AToliasLab @danilobzdok
-
-
Odgovor korisnicima @neuro_data @AstroKatie i sljedećem broju korisnika:
I think this is a weird phrasing. You can have an unsupervised learning algorithm that leads to representations that are better for downstream RL or supervised tasks. The purpose of the unsupervised alg is indeed beteer downstream performance, but the alg is still unsupervised.
0 proslijeđenih tweetova 5 korisnika označava da im se sviđa -
Odgovor korisnicima @tyrell_turing @AstroKatie i sljedećem broju korisnika:
yes, that is my point that it is weird phrasing. you are a tiger and you just attached a rhino and didn't kill it. now, you are watching the rhino, without "labeled data". why? to build a better model *that enables you to act better in the future attack*.
1 reply 0 proslijeđenih tweetova 1 korisnik označava da mu se sviđa -
Odgovor korisnicima @neuro_data @tyrell_turing i sljedećem broju korisnika:
i guess my point is: calling it unsupervised is a bit misleading, it has a specific goal, and it uses past training data to get better at that goal, not simply "learn a parsimonious model" for its own sake.
0 proslijeđenih tweetova 2 korisnika označavaju da im se sviđa -
Odgovor korisnicima @neuro_data @AstroKatie i sljedećem broju korisnika:
True, but the reason I would still call it "unsupervised" is because the learning algorithm itself is using only the sensory data, no other rewards or external targets, to determine the parameter updates.
0 proslijeđenih tweetova 7 korisnika označava da im se sviđa -
Odgovor korisnicima @tyrell_turing @neuro_data i sljedećem broju korisnika:
You can break this down into a few possibilities 1) something like this, which is not simply optimizing some parsimonious/simple objective like “compression” or “reconstruction”, and whose goal is downstream task performance, but is still unsupervised: https://arxiv.org/abs/1804.00222
0 proslijeđenih tweetova 1 korisnik označava da mu se sviđa -
Odgovor korisnicima @AdamMarblestone @tyrell_turing i sljedećem broju korisnika:
2) unsupervised, e.g., predictive, learning to keep certain representations around, which then happens to make it easier to learn from external rewards https://arxiv.org/abs/1803.10760
1 reply 0 proslijeđenih tweetova 1 korisnik označava da mu se sviđa
3) my favorite, highly specific internal loss functionspic.twitter.com/X0Zc65LdvV
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.