Artificial Neural Networks as Non-Linear Extensions of Statistical Methods in Astronomy

Abstract

We attempt to de-mistify Artificial Neural Networks (ANNs) by considering special cases which are related to other statistical methods common in Astronomy and other fields. In particular we show how ANNs generalise Bayesian methods, multi-parameter fitting, Principal Component Analysis (PCA), Wiener filtering and regularisation methods. Examples of morphological classification of galaxies illustrate how non-linear ANNs improve on linear techniques.

9 pages, uu-encoded compressed postscript file. Also available by anonymous ftp to cast0.ast.cam.ac.uk (131.111.68.35) at ftp://cast0.ast.cam.ac.uk/pub/lahav/vistas/vistas4.ps.Z with figure at ftp://cast0.ast.cam.ac.uk/pub/lahav/vistas/fig1.ps.Z To appear in Vistas in Astronomy, special issue on Artificial Neural Networks in Astronomy

References in corpus (2)

• Galaxies, Human Eyes and Artificial Neural Networks
• Wiener Reconstruction of Galaxy Surveys in Spherical Harmonics

Cited by in corpus (4)

• Astronomia ex machina: a history, primer, and outlook on neural networks in astronomy
• Automated Classification of 2000 Bright IRAS Sources
• Deep Learning Jet Image as a Probe of Light Higgsino Dark Matter at the LHC
• Augmenting photometric redshift estimates using spectroscopic nearest neighbours