pretraining isn't unsupervised, it is self-supervised - meaning it is moderately more scale limited.

sigbottle 10 hours ago | parent | next [-]

What would unsupervised mean, would unsupervised be something like alphago playing against itself trillions of times?

Whereas self-supervised, allows learning without explicit annotation of data ; but it doesn't matter if the models already trained on the entire Internet, and it's not like a game where it can come up with effectively new training data for itself?

	▲	jmalicki 6 hours ago \| parent [-]
		Unsupervised is basically clustering. Alphago is RL - winning or losing a game is a form of supervision. Unsupervised is something where there is no intrinsic reward signal. In pre training, predicting the next token and seeing that it matches is a reward signal, hence it is self supervised.

▲

cold_harbor 9 hours ago | parent | prev [-]

fair point — OpenAI's original plan literally said "solve unsupervised learning". the self-supervised distinction wasnt really standard til after BERT/GPT popularized it

	▲	jmalicki 5 hours ago \| parent [-]
		I think it's an extremely important distinction because self supervised learning has real inherent reward signals. Something like clustering does not.