Deep one-gate per layer networks with skip connections are universal classifiers

TL;DR

This paper demonstrates that deep neural networks with one-gate layers and skip connections can be constructed from simpler multilayer perceptrons, highlighting their potential as universal classifiers.

Contribution

It introduces a method to transform two-hidden-layer perceptrons into deep networks with one-gate layers and skip connections, emphasizing their universality.

Findings

Deep networks with one-gate layers and skip connections can classify data effectively.

Transformation from simple perceptrons to deep networks is straightforward.

The approach shows the universality of such deep architectures.

Abstract

This paper shows how a multilayer perceptron with two hidden layers, which has been designed to classify two classes of data points, can easily be transformed into a deep neural network with one-gate layers and skip connections.