Poster
Transition to Linearity of General Neural Networks with Directed Acyclic Graph Architecture
Libin Zhu · Chaoyue Liu · Misha Belkin
Hall J (level 1) #934
Keywords: [ transition to linearity ] [ over-parameterization ] [ Neural Tangent Kernel ] [ wide neural networks ] [ Directed Acyclic Graph ]
In this paper we show that feedforward neural networks corresponding to arbitrary directed acyclic graphs undergo transition to linearity as their ``width'' approaches infinity. The width of these general networks is characterized by the minimum in-degree of their neurons, except for the input and first layers. Our results identify the mathematical structure underlying transition to linearity and generalize a number of recent works aimed at characterizing transition to linearity or constancy of the Neural Tangent Kernel for standard architectures.