AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Is Linearity Dominating Machine Learning?
In the last decade, we've also seen reus come to dominance in the field of obneral networks. A recent work by randall a bellistriero developed an interesting frame of reference which casts multi layer perceptrones as a decomposition method. So my question is, why is linearity, whether it's pieace wise or otherwise, dominating the state of the art and approximation methods? It almost seems to me like we'v kind of gone back to the future, if you will,. sort of leaving behind attempts at more smooth, non linear methods and gone back to newer, albeit more complicated, forms of of linear approximation. Right? Veri col yes, thank you