caa a22 4500 477073042 CHVBK 20180405111431.0 cr unu---uuuuu 170330e19960601xx s 000 0 eng 10.1007/BF01413747 doi (NATIONALLICENCE)springer-10.1007/BF01413747 Use of methodological diversity to improve neural network generalisation [Elektronische Daten] [W. Yates, D. Partridge] Littlewood and Miller [4] present a statistical framework for dealing with coincident failures in multiversion software systems. They develop a theoretical model that holds the promise of high system reliability through the use of multiple, diverse sets of alternative versions. In this paper, we adapt their framework to investigate the feasibility of exploiting the diversity observable in multiple populations of neural networks developed using diverse methodologies. We evaluate the generalisation improvements achieved by a range of methodologically diverse network generation processes. We attempt to order the constituent methodological features with respect to their potential for use in the engineering of useful diversity. We also define and explore the use of relative measures of the diversity between version sets as a guide to the potential for exploiting interset diversity. Springer-Verlag London Limited, 1996 Multilayer perceptrons nationallicence Backpropagation nationallicence Radial basis function networks nationallicence Generalisation nationallicence Multiversion systems nationallicence Software reliability nationallicence Yates W. Department of Computer Science, University of Exeter, EX4 4PT, Exeter, UK aut Partridge D. Department of Computer Science, University of Exeter, EX4 4PT, Exeter, UK aut Neural Computing & Applications Springer-Verlag 4/2(1996-06-01), 114-128 0941-0643 4:2<114 1996 4 521 https://doi.org/10.1007/BF01413747 text/html Onlinezugriff via DOI 1 research-article jats NATIONALLICENCE 856 40 https://doi.org/10.1007/BF01413747 text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Yates W. Department of Computer Science, University of Exeter, EX4 4PT, Exeter, UK aut NATIONALLICENCE 700 1- Partridge D. Department of Computer Science, University of Exeter, EX4 4PT, Exeter, UK aut NATIONALLICENCE 773 0- Neural Computing & Applications Springer-Verlag 4/2(1996-06-01), 114-128 0941-0643 4:2<114 1996 4 521 Metadata rights reserved Springer special CC-BY-NC licence nationallicence BK010053 XK010053 XK010000 NATIONALLICENCE NATIONALLICENCE NL-springer