Controlling network ensembles

Analysis of Network Ensembles

Neural Network Ensembles

We propose several means for improving the performance and training of neural networks for classification. We use crossvalidation as a tool for optimizing network parameters and architecture. We show further that the remaining residual “generalization” error can be reduced by invoking ensembles of similar networks. Zndex Terms-Crossvalidation, fault tolerant computing, neural networks, N-versio...

Immune Network based Ensembles

This paper presents a new method for constructing ensembles of classifiers based on immune network theory, one of the most interesting paradigms within the field of artificial immune systems. Ensembles of classifiers are a very interesting alternative to single classifiers when facing difficult problems. In general, ensembles are able to achieve better performance in terms of learning and gener...

Structural Trends in Network Ensembles

A collection of networks is considered a network ensemble if its members originate from a common natural or technical process such as repeated measurements, replication and mutation, or massive parallelism, possibly under varying conditions. We propose a spectral approach to identify structural trends, i. e. prevalent patterns of connectivity, in an ensemble by delineating classes of networks w...

Unstable Communities in Network Ensembles

Ensembles of graphs arise in several natural applications. Many techniques exist to compute frequent, dense subgraphs in these ensembles. In contrast, in this paper, we propose to discover maximally variable regions of the graphs, i.e., sets of nodes that induce very different subgraphs across the ensemble. We first develop two intuitive and novel definitions of such node sets, which we then sh...