Tosh, C. R. and Ruxton, G. D. (2007) The need for stochastic replication of ecological neural networks. Philosophical Transactions of the Royal Society B: Biological Sciences, 362(1479), pp. 455-460. (doi: 10.1098/rstb.2006.1973) (PMID:17255011) (PMCID:PMC2323564)
Text
pubmedredirect.html 4kB |
Abstract
Artificial neural networks are becoming increasingly popular as predictive statistical tools in ecosystem ecology and as models of signal processing in behavioural and evolutionary ecology. We demonstrate here that a commonly used network in ecology, the three-layer feed-forward network, trained with the backpropagation algorithm, can be extremely sensitive to the stochastic variation in training data that results from random sampling of the same underlying statistical distribution, with networks converging to several distinct predictive states. Using a random walk procedure to sample error–weight space, and Sammon dimensional reduction of weight arrays, we demonstrate that these different predictive states are not artefactual, due to local minima, but lie at the base of major error troughs in the error–weight surface. We further demonstrate that various gross weight compositions can produce the same predictive state, suggesting the analogy of weight space as a ‘patchwork’ of multiple predictive states. Our results argue for increased inclusion of stochastic training replication and analysis into ecological and behavioural applications of artificial neural networks.
Item Type: | Articles |
---|---|
Status: | Published |
Refereed: | Yes |
Glasgow Author(s) Enlighten ID: | Ruxton, Professor Graeme and Tosh, Dr Colin |
Authors: | Tosh, C. R., and Ruxton, G. D. |
Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QP Physiology |
College/School: | College of Medical Veterinary and Life Sciences > School of Biodiversity, One Health & Veterinary Medicine |
Journal Name: | Philosophical Transactions of the Royal Society B: Biological Sciences |
Publisher: | The Royal Society |
ISSN: | 0962-8436 |
ISSN (Online): | 1471-2970 |
University Staff: Request a correction | Enlighten Editors: Update this record