AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Biggest Argument Against the Predictive Task
Soha: The structure of the laten table one is the biggest am the biggest argument against, the most prevalent argument against this paper. We almost need to have an information theoretical, theoretic way of looking at this. But anyway, i might my intuition is that deep lonin models, in code the most high frequency information into the latent spaceright? And you know, this information would be incoded in a minimally distributed way to denois the predictive task.