MODULARIZING DEEP LEARNING VIA PAIRWISE LEARNING WITH KERNELS

 By redefining the conventional notions of layers, we present an alternative view on finitely wide, fully trainable deep neural networks as stacked linear models in feature spaces, leading to a kernel machine interpretation. Based on this construction, we then propose a provably optimal modular learning framework for classification that does not req...

DEEP LEARNING MODULARIZING DEEP LEARNING

MODULARIZING DEEP LEARNING VIA PAIRWISE LEARNING WITH KERNELS

 By redefining the conventional notions of layers, we present an alternative view on finitely wide, fully trainable deep neural networks as stacked linear models in feature spaces, leading to a kernel machine interpretation. Based on this construction, we then propose a provably optimal modular learning framework for classification that does not req...

DEEP LEARNING VIA PAIRWISE LEARNING