Designer Algorithms for Astronomy
Large archives of
astronomical data (images, spectra and catalogues) are being assembled into a
database that will soon be accessible worldwide as part of the
Virtual Observatory.
This necessitates the development of techniques that will allow fast, automated
classification and extraction of key physical properties for very large
datasets, and the ability to visualise the structure of highly multi-dimensional
data, and extract and study substructures in a flexible way. This collaboration
between the
School of Physics and Astronomy and the
School of
Computer Science at the University of Birmingham aims to adapt, for
the use of astronomical data mining, a number of innovative algorithms for data
analysis including
genetic programming and
evolutionary computation, latent variable analysis,
computer vision,
machine learning networks
etc.
The
effort is coordinated by
Somak
Raychaudhury, and involves both astronomers (Trevor
Ponman, Louisa Nolan, Ian Stevens, Bill Chaplin, Habib Khosroshahi) and
Computer Scientists
(Ela Claridge,
Ata Kaban,
Alan Sexton,
Peter Tino,
and Xin Yao),
at Birmingham.
This activity
recently led to a E-science award from PPARC, as a result of which
Jianyong Sun has been working as a
postdoctoral fellow in this field.
-
Latent
variable and Bayesian modelling-
the use of Kernel methods and Support Vector
machines: Measuring time-delays
between multiple images in a gravitational lens, using long-term monitoring
with high-resolution radio or optical images, is difficult in the presence of
correlated noise on various scales. We are investigating various methods of
doing this. Bayesian methods developed with Markus Harva (Helsinki) have
met with some success, and a kernel-fitting approach using latent
variables is the subject of a PhD thesis (Juan
Cuevas Tello), jointly supervised by Dr Peter Tino (Computer Science) and
Somak Raychaudhury (ASR group). In principle, such modelling could lead us
to measure time-delayed signals from unresolved images. This method has much
wider application, which will be our next goal, in applying it to
automatically finding redshifts from millions of galaxy spectra and
quantifying the spectral widths of emission and absorption lines.
-
Independent Component analysis of
galaxy spectra: Elliptical galaxies were once
believed to consist of a single population of old stars formed coevally at
high redshift, followed by predominantly passive evolution. However, more
recent hierarchical structure formation models suggest that they are formed
from the low redshift merging of disk galaxies, with associated significant
star formation, and recent analyses of galaxy spectra seem to indicate the
presence of significant younger populations of stars in at least some
elliptical galaxies. The detailed physical modelling of such populations via
spectral fitting, is computationally expensive, inhibiting the detailed
analysis of the several million galaxy spectra which will become available
over the next few years. A collaboration between Ata Kaban, Markus Harva, Louisa
Nolan and Somak Raychaudhury has developed a data-driven application aimed at decomposing the
spectra of galaxies into that of several stellar populations, without the use
of detailed physical models. This method includes a Bayesian way of filling in
missing data in an ensemble of spectra, and the interpretation of the
independent components is terms of old and young stellar populations has
already yielded spectacular results.
-
Hierarchical visualization of high dimensional data: Jianyong Sun, Ata Kaban, Peter Tino, Somak Raychaudhury (more
to come)
-
Inversion
techniques for spectral mapping: An inversion technique for the recovery
of physical parameters from multi-colour images, already successfully applied
in medical imaging, has been applied to X-ray images, extracted in a
set of optimal energy bands, to map the spectral properties of hot gas in
clusters of galaxies. This effort,
resulting from a collaboration between Ela Claridge, Mark O'Dwyer (CS),
Trevor Ponman
and Somak Raychaudhury, will facilitate extensive statistical studies of physical
properties of galaxies and clusters from large X-ray archives without detailed
model-fitting.
-
Genetic algorithms for model
discovery: As data improve, the analytical
forms traditionally used to model galaxies and their clusters prove
to be inadequate, where departures from such simple forms may
contain important information on structure and evolution. To provide
a more flexible and sophisticated suite of models, developed by Jin
Li, Xin Yao (CS), Habib Khosroshahi and Somak Raychaudhury, we
are examining the use of genetic algorithms, which allow models
themselves to evolve, in a fashion modelled on biological evolution,
to fit photometric observations of both galaxies and clusters.
Publications in this field
Refereed Journals
-
Finding young stellar populations
in early-type galaxies from an independent component analysis of
their UV-optical spectra
Nolan Louisa A., Harva Markus O.,
Kaban Ata and Raychaudhury Somak, 2006, MNRAS, 366, 321-338.
- How accurate are the time delay
estimates in gravitational lensing?
Juan C. Cuevas-Tello, Peter Tino& Somak Raychaudhury 2006,
Astron. Astrophys., 454, 695-706
-
Young stellar populations in early-type galaxies in the Sloan Digital Sky Survey
Nolan Louisa A., Raychaudhury Somak
and Kaban Ata, 2006, MNRAS, submitted.
(astro-ph/0608623)
- A Bayesian approach to estimating time
delays between gravitationally lensed multiple images
Harva Markus O. and Raychaudhury Somak,
2006, MNRAS, submitted.
Refereed Conference Proceedings
-
An Evolutionary Approach
to Modelling Radial Brightness Distributions in Elliptical Galaxies
J. Li, X. Yao, C. Frayn, H. G. Khosroshahi and S. Raychaudhury, 2004,
Lecture Notes in Computer Science 3242, 591-601. Berlin: Springer-Verlag. (ISBN:3540230920),
Proceedings of the 8th
International Conference on Parallel Problem Solving from Nature (PPSN
VIII), September 2004)
-
Mapping the physical
properties of cosmic hot gas with hyper-spectral imaging
Ela Claridge, Mark O'Dwyer, Trevor Ponman, and Somak Raychaudhury,
2005, in IEEE Workshop on Applications
of Computer Vision (WACV2005), January 2005, pp 185-190.
(astro-ph/0505165)
- Finding Young
Stellar Populations in Elliptical Galaxies from Independent Components of
Optical Spectra
Ata Kaban, Louisa Nolan, Somak Raychaudhury, 2005,
SIAM International Conference on Data Mining (SDM05)
21--24 April 2005, Newport Beach California, USA.
(astro-ph/0505059)
-
Bayesian Estimation of Time Delays Between Unevenly Sampled Signals
Markus Harva and Somak Raychaudhury, 2006, accepted by
IEEE International Workshop on Machine Learning for signal processing,
September 2006, Maynooth, Ireland
- A kernel-based approach to estimating phase shifts between irregularly sampled time series: an application to gravitational lenses
Juan C. Cuevas-Tello, Peter Tino and Somak Raychaudhury, 2006, accepted by
The 17th European Conference on Machine Learning (ECML06),
September 2006, Berlin
-
Factorisation of positive valued functions
Ata Kaban, Louisa Nolan and Somak Raychaudhury, 2006, accepted by
ICA Research Network International Workshop (ICArn06),
September 2006, Liverpool
-
On class visualisation for high dimensional data: Exploring scientific data sets
Ata Kaban, Jianyong Sun, Somak Raychaudhury and
Louisa Nolan, 2006, accepted by Ninth International conference on Discovery Science (DS-2006),
October 2006, Barcelona, Spain (Publisher: Springer)
Un-refereed Conference Proceedings and Posters
-
Time delay estimation in gravitational lensing: a new approach
Cuevas-Tello Juan C., Raychaudhury, Somak, Tino Peter, 2005,
abstract in the proceedings of the RAS National Astronomy Meeting,
Birmingham, April 2005.
-
Finding young stellar populations in early-type galaxies from independent components of their UV-optical spectra
Nolan, L., Kaban, A., Raychaudhury, Somak, 2005,
abstract in the proceedings of the RAS National Astronomy Meeting,
Birmingham, April 2005.
-
Determining time delay in gravitational lending: how significant are the results?
Tino, Peter, Cuevas-Tello Juan C., Raychaudhury, Somak, 2005,
abstract in the proceedings of the RAS National Astronomy Meeting,
Birmingham, April 2005.
-
Kernel-based methods applied to irregularly sampled time series
Cuevas-Tello Juan C., Tino, Peter,
Raychaudhury, Somak, 2005, Poster in
"The Analysis of Patterns",
Centre Ettore Majorana for Scientific Culture, Erice, Italy
October 28 - November 6, 2005.
|