RNN

Recurrent Neural Network

Neural net for sequences -- mostly superseded by transformers for language.

When you'd see it: Background on how sequence models worked before transformers. Still seen in some time-series and signal applications.

Why it matters: RNNs process a sequence step by step, carrying a memory of what came before, which made them the early workhorse for language and time-series. Transformers have largely replaced them for language, but the concept explains the lineage.