AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Harder Learning Problem
In those days, we were still back solidly in the classic NLP pipeline kind of way of thinking about things. So there happened a little bit of machine learning for this, but you would kind of require labeled data of all of those steps. And at the time, latent variable learning problems with, you know, unlabeled variables where you do marginal inference and things like that were all the rage. It was kind of well timed to you and there's a lot of interest in that area.