Abstract
We describe a framework for adaptive astronomical exploration based on iterating an Observation-Inference-Design cycle that allows adjustment of hypotheses and observing protocols in response to the results of observation on-the-fly, as data are gathered. The framework uses a unified Bayesian methodology for the inference and design stages: Bayesian inference to quantify what we have learned from the available data; and Bayesian decision theory to identify which new observations would teach us the most. In the design stage, the utility of possible future observations is determined by how much information they are expected to add to current inferences as measured by the (negative) entropies of the probability distributions involved. Such a Bayesian approach to experimental design dates back to the 1970s, but most existing work focuses on linear models. We use a simple nonlinear problem—planning observations to best determine the orbit of an extrasolar planet—to illustrate the approach and demonstrate that it can significantly improve observing efficiency (i.e., reduce uncertainties at a rate faster than the familiar “root-N” rule) in some situations. We highlight open issues requiring further research, including dependence on model specification, generalizing the utility of an observation (e.g., to include observing “costs”), and computational issues.
This paper is followed by a commentary by David A. van Dyk.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
A.C. Atkinson and V. V. Fedorov. Optimum design of experiments. In Samuel Kotz, Campbell B. Read, and David L. Banks, editors, Encyclopedia of statistical sciences, pages 107–114. Wiley, New York, 1997.
H. Cherno. Sequential Analysis and Optimal Design. SIAM, Philadelphia, 1972.
M. Clyde, P. Muller, and G. Parmigiani. Exploring expected utility surfaces by markov chains, 1995.
K. Chaloner and I. Verdinelli. Bayesian experimental design: A review. Stat. Sci., 10:273–304, 1995.
V. V. Fedorov. Theory of Optimal Experiments. Academic, New York, 1972.
W. L. Freedman, S. M. Hughes, B. F. Madore, J. R. Mould, M. G. Lee, P. Stetson, R. C. Kennicutt, A. Turner, L. Ferrarese, H. Ford, J. A. Graham, R. Hill, J. G. Hoessel, J. Huchra, and G. D. Illingworth. The hubble space telescope extragalactic distance scale key project. 1: The discovery of cepheids and a new distance to m81. Ap. J., 427:628–655, June 1994.
D. V. Lindley. On the measure of information provided by an experiment. Ann. Stat., 27:986–1005, 1956.
D. V. Lindley. Bayesian statistics-a review. SIAM, Philadelphia, 1972.
T. J. Loredo. Computational technology for bayesian inference. In ASP Conf. Ser. 172: Astronomical Data Analysis Software and Systems VIII, volume 8, pages 297+, 1999.
D. MacKay. Information-based objective functions for active data selection. Neural Computation, 4 (4):590–604, 1992.
P. Muller and G. Parmigiani. Numerical evaluation of information theoretic measures. In D. A. Berry, K. M Chaloner, and J. F. Geweke, editors, Bayesian Statistics and Econometrics: Essays in Honor of A. Zellner, pages 397–406. Wiley, New York, 1995.
Peter Muller and Giovanni Parmigiani. Optimal design via curve fitting of monte carlo experiments. J. Am. Stat. Assoc., 90:1322–1330, 1996.
P. Muller. Simulation based optimal design. In J. O. Berger, J. M. Bernardo, A. P. Dawid, and A. F. M. Smith, editors, Bayesian Statistics 6, 1999.
W. H. Press, S. A. Teukolsky, W. T. Vetterling, and B. P. Flannery. Numerical Recipes in C: The Art of Scientific Computing. Cambridge University Press, Cambridge, 1992.
P. Sebastiani and R. Settimi. First-order optimal designs for non-linear models. J. Stat. Plan. Inf., 74:177–192, 1998.
P. Sebastiani and H. P. Wynn. Bayesian experimental design and shannon information. In 1997 Proceedings of the Section on Bayesian Statistical Science, pages 176–181. American Statistical Association, 1997.
P. Sebastiani and H. P. Wynn. Maximum entropy sampling and optimal bayesian experimental design. J. Roy. Stat. Soc. B, 62:145–157, 2000.
B. Toman. Bayesian experimental design for multiple hypothesis testing. J. Am. Stat. Assoc., 91:185–190, 1996.
B. Toman. Bayesian experimental design. In Samuel Kotz and N. L. Johnson, editors, Encyclopedia of statistical sciences, Update Vol. 3, pages 35–39. Wiley, New York, 1999.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag New York, Inc.
About this paper
Cite this paper
Loredo, T.J., Chernoff, D.F. (2003). Bayesian Adaptive Exploration. In: Statistical Challenges in Astronomy. Springer, New York, NY. https://doi.org/10.1007/0-387-21529-8_4
Download citation
DOI: https://doi.org/10.1007/0-387-21529-8_4
Publisher Name: Springer, New York, NY
Print ISBN: 978-0-387-95546-9
Online ISBN: 978-0-387-21529-7
eBook Packages: Springer Book Archive