Remix clone Hacker News

new | show | ask | jobs Github

▲

drdeca 7 days ago

Why are you conflating MLPs in general with specifically transformers?

▲

quantadev 7 days ago | parent [-]

I consider MLPs the building blocks of all this, and is what makes things a neural net, as opposed to some other data structure.

▲

drdeca 7 days ago | parent [-]

Sure. But that isn’t a reason to conflate the two?

OP wasn’t suggesting looking for an alternative/successor to MLPs, but for an alternative/successor to transformers (while presumably still using MLPs) in the same way that transformers are an alternative/successor to LSTMs.

	▲	quantadev 7 days ago \| parent [-]
		And that sort of proves my original point which is that we're probably gonna keep riding the same wave as far as it will go!! i.e. keep the tech stack mostly with just what we know works best.