Twitter | Pretraživanje | |
jovo 8. pro
here’s a conjecture: biological learning is never unsupervised. Rather, its goal is to learn representations that are useful for future behavior.
Reply Retweet Označi sa "sviđa mi se"
Blake Richards 8. pro
Odgovor korisniku/ci @neuro_data @AstroKatie i 2 ostali
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.
Reply Retweet Označi sa "sviđa mi se"
jovo 9. pro
Odgovor korisniku/ci @tyrell_turing @AstroKatie i 2 ostali
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*.
Reply Retweet Označi sa "sviđa mi se"
jovo 9. pro
Odgovor korisniku/ci @tyrell_turing @AstroKatie i 2 ostali
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.
Reply Retweet Označi sa "sviđa mi se"
Blake Richards 9. pro
Odgovor korisniku/ci @neuro_data @AstroKatie i 2 ostali
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.
Reply Retweet Označi sa "sviđa mi se"
Adam Marblestone 9. pro
Odgovor korisniku/ci @tyrell_turing @neuro_data i 3 ostali
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:
Reply Retweet Označi sa "sviđa mi se"
Adam Marblestone 9. pro
Odgovor korisniku/ci @tyrell_turing @neuro_data i 3 ostali
2) unsupervised, e.g., predictive, learning to keep certain representations around, which then happens to make it easier to learn from external rewards
Reply Retweet Označi sa "sviđa mi se"