On Minimal Depth in Neural Networks

arXiv:2402.15315v1 Announce Type: new
Abstract: A characterization of the representability of neural networks is relevant to comprehend their success in artificial intelligence. This study investigate two topics on ReLU neural network expressivity and their connection with a conjecture related to the minimum depth required for representing any continuous piecewise linear function (CPWL). The topics are the minimal depth representation of the sum and max operations, as well as the exploration of polytope neural networks. For the sum operation, we establish …

abstract artificial artificial intelligence arxiv conjecture continuous cs.dm cs.lg function intelligence linear math.co network networks neural network neural networks relu study success topics type