SplitNet: A Dynamic Hierarchical Network Model

Abstract

Introduction In this work, we investigate the information that is contained in the structure of a topology preserving neural network. We consider a topological map as a graph G, propose certain properties of the structure and formulate the respective expectable results of network interpretation. From the results we conclude that topology preservation as well as neuron distribution are highly influential for the network semantics. Guided by these insights, we developed a new network model that fits both needs. This so called SplitNet model dynamically constructs a hierarchically structured network that provides interpretability by neuron distribution, network topology and hierarchy of the network layers. The scenario we deal with is the nearest-neighbor approach to classification. The problems are to find the number and positions of neurons that is useful and efficient for the given data and to retrieve a list L of m nearest neigh