×

Learning functions and approximate Bayesian computation design: ABCD. (English) Zbl 1341.62091

Summary: A general approach to Bayesian learning revisits some classical results, which study which functionals on a prior distribution are expected to increase, in a preposterior sense. The results are applied to information functionals of the Shannon type and to a class of functionals based on expected distance. A close connection is made between the latter and a metric embedding theory due to Schoenberg and others. For the Shannon type, there is a connection to majorization theory for distributions. A computational method is described to solve generalized optimal experimental design problems arising from the learning framework based on a version of the well-known approximate Bayesian computation (ABC) method for carrying out the Bayesian analysis based on Monte Carlo simulation. Some simple examples are given.

MSC:

62F15 Bayesian inference
62K05 Optimal statistical designs
94A17 Measures of information, entropy
PDFBibTeX XMLCite
Full Text: DOI

References:

[2] Torgersen, Comparison of Statistical Experiments (1991)
[4] Lindley, On a Measure of the Information Provided by an Experiment, Ann. Math. Stat 27 pp 986– (1956) · Zbl 0073.14103 · doi:10.1214/aoms/1177728069
[5] Goel, Comparison of Experiments and Information Measures, Ann. Math. Stat 7 pp 1066– (1979) · Zbl 0412.62004 · doi:10.1214/aos/1176344790
[6] Ginebra, On the measure of the information in a statistical experiment, Bayesian Anal 2 pp 167– (2007) · Zbl 1331.62056 · doi:10.1214/07-BA207
[7] Chaloner, Bayesian Experimental Design: A Review, Stat. Sci 10 pp 273– (1995) · Zbl 0955.62617 · doi:10.1214/ss/1177009939
[8] Sebastiani, Maximum entropy sampling and optimal Bayesian experimental design, J. R. Stat. Soc.: Ser. B (Stat. Methodol.) 62 pp 145– (2000) · Zbl 0941.62084 · doi:10.1111/1467-9868.00225
[9] Chater, The Probability Heuristics Model of Syllogistic Reasoning, Cogn. Psychol 38 pp 191– (1999) · Zbl 02332493 · doi:10.1006/cogp.1998.0696
[10] Schoenberg, Metric Spaces Positive Definite Functions, Trans. Am. Math. Soc 44 pp 522– (1938) · JFM 64.0617.02 · doi:10.1090/S0002-9947-1938-1501980-0
[11] Schilling, Bernstein Functions: Theory and Applications (2012)
[12] Tsallis, Possible generalization of Boltzmann-Gibbs statistics, J. Stat. Phys 52 pp 479– (1988) · Zbl 1082.82501 · doi:10.1007/BF01016429
[13] DeGroot, Optimal Statistical Decisions (2004)
[14] Goldman, Results on inquiry and truth possession, Stat. Probab. Lett 12 pp 415– (1991) · Zbl 0742.62108 · doi:10.1016/0167-7152(91)90030-U
[15] Fallis, Further results on inquiry and truth possession, Stat. Probab. Lett 60 pp 169– (2002) · Zbl 1014.62001 · doi:10.1016/S0167-7152(02)00308-5
[16] Torgerson, Theory and Methods of Scaling (1958)
[17] Gower, Some distance properties of latent root and vector methods used in multivariate analysis, Biometrika 53 pp 325– (1966) · Zbl 0192.26003 · doi:10.1093/biomet/53.3-4.325
[18] Gower, Euclidean distance geometry, Math. Sci 7 pp 1– (1982) · Zbl 0492.51017
[19] Itti, Bayesian surprise attracts human attention, Vis. Res 49 pp 1295– (2009) · doi:10.1016/j.visres.2008.09.007
[20] Haykin, The Cocktail Party Problem, Neural Comput 17 pp 1875– (2005) · Zbl 02231094 · doi:10.1162/0899766054322964
[21] Berger, The case for objective Bayesian analysis, Bayesian Anal 1 pp 385– (2006) · Zbl 1331.62042 · doi:10.1214/06-BA115
[22] Marshall, Inequalities: Theory of Majorization and Its Applications (2009)
[23] Hardy, Inequalities (1988)
[24] Müller, Comparison Methods for Stochastic Models and Risks (2002) · Zbl 0999.60002
[25] Ryff, Orbits of l1-functions under doubly stochastic transformations, Trans. Am. Math. Soc 117 pp 92– (1965) · Zbl 0135.18804
[26] DeGroot, Comparing probability forecasters: Basic binary concepts and multivariate extensions, Bayesian Inference and Decision Techniques pp 247– (1986)
[27] Dawid, Coherent dispersion criteria for optimal experimental design, Ann. Stat 27 pp 65– (1999) · Zbl 0948.62057
[28] Marjoram, Markov chain Monte Carlo without likelihoods, Proc. Natl. Acad. Sci. USA 100 pp 15324– (2003) · doi:10.1073/pnas.0306899100
[29] Hainy, Approximate Bayesian Computation Design (ABCD), an Introduction, mODa 10–Advances in Model-Oriented Design and Analysis pp 135– (2013)
[30] Drovandi, Bayesian Experimental Design for Models with Intractable Likelihoods, Biom 69 pp 937– (2013) · Zbl 1419.62182 · doi:10.1111/biom.12081
[31] Hainy, Likelihood-free Simulation-based Optimal Design (2013)
[32] Müller, Beyond space-filling: An illustrative case, Procedia Environ. Sci 7 pp 14– (2011) · doi:10.1016/j.proenv.2011.07.004
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.