×

Gaussian processes for computer experiments. (English. French summary) Zbl 1388.60082

Summary: This paper collects the contributions which were presented during the session devoted to Gaussian processes at the Journées MAS 2016. First, an introduction to Gaussian processes is provided, and some current research questions are discussed. Then, an application of Gaussian process modeling under linear inequality constraints to financial data is presented. Also, an original procedure for handling large data sets is described. Finally, the case of Gaussian process based iterative optimization is discussed.

MSC:

60G15 Gaussian processes
91G80 Financial applications of other theories
91B70 Stochastic models in economics
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] P. Abrahamsen. A review of Gaussian random fields and correlation functions. Technical report, Norwegian computing center,1997.
[2] R.J. Adler. {\it An introduction to continuity, extrema, and related topics for general Gaussian processes}. Hayward, CA: Instituteof mathematical statistics, 1990. · Zbl 0747.60039
[3] P. Auer, N. Cesa-Bianchi, and P. Fischer. Finite-time analysis of the multiarmed bandit problem. {\it Machine Learning}, 47(2-3):235-256, 2002. · Zbl 1012.68093
[4] F. Bachoc. Cross validation and maximum likelihood estimations of hyper-parameters of Gaussian processes with modelmispecification. {\it Computational Statistics and Data Analysis}, 66:55-69, 2013. · Zbl 1471.62021
[5] F. Bachoc. Asymptotic analysis of the role of spatial sampling for covariance parameter estimation of Gaussian processes.{\it Journal of Multivariate Analysis}, 125:1-35, 2014. · Zbl 1280.62100
[6] F. Bachoc. Asymptotic analysis of covariance parameter estimation for Gaussian processes in the misspecified case. {\it Accepted}{\it in Bernoulli}, 2016. · Zbl 1429.60035
[7] F. Bachoc, K. Ammar, and J.M. Martinez. Improvement of code behavior in a design of experiments by metamodeling. {\it Nuclear}{\it science and engineering}, 183(3):387-406, 1016.
[8] F. Bachoc, G. Bois, J. Garnier, and J.M Martinez. Calibration and improved prediction of computer models by universalKriging. {\it Nuclear Science and Engineering}, 176(1):81-97, 2014.
[9] F. Bachoc, N. Durrande, D. Rullière, and C. Chevalier. Some properties of nested Kriging predictors. {\it arXiv preprint}{\it arXiv:1707.05708}, 2017. · Zbl 1495.86009
[10] P. Billingsley. {\it Probability and Measure, 3rd edition}. Wiley, New York, 1995. · Zbl 0822.60002
[11] A. D. Bull. Convergence rates of efficient global optimization algorithms. {\it The Journal of Machine Learning Research}, 12:2879-2904, 2011. · Zbl 1280.90094
[12] E. Contal. {\it Statistical learning approaches for global optimization}. PhD thesis, Université Paris-Saclay, 2016.
[13] E. Contal and N. Vayatis. Stochastic process bandits: Upper confidence bounds algorithms via generic chaining. {\it arXiv preprint}{\it arXiv:1602.04976}, 2016.
[14] A. Cousin, H. Maatouk, and D. Rullière. Kriging of financial term-structures. {\it European Journal of Operational Research},255(2):631 - 648, 2016. · Zbl 1346.91264
[15] A. Cousin and I. Niang. On the range of admissible term-structures. {\it arXiv preprint arXiv:1404.0340}, 2014.
[16] H. Cramer and R. Leadbetter. {\it Stationary and related stochastic processes: sample function properties and their applications}.Wiley series in probability and mathematical statistics. Tracts on probability and statistics. 1967.178ESAIM: PROCEEDINGS AND SURVEYS · Zbl 0162.21102
[17] N. Cressie and G. Johannesson. Fixed rank Kriging for very large spatial data sets. {\it Journal of the Royal Statistical Society:}{\it Series B (Statistical Methodology)}, 70(1):209-226, 2008. · Zbl 05563351
[18] N. Cressie and S.N. Lahiri. The asymptotic distribution of REML estimators. {\it Journal of Multivariate Analysis}, 45:217-233,1993. · Zbl 0772.62008
[19] N. Cressie and S.N. Lahiri. Asymptotics for REML estimation of spatial covariance parameters. {\it Journal of Statistical Planning}{\it and Inference}, 50:327-341, 1996. · Zbl 0847.62044
[20] S. Da Veiga and A. Marrel. Gaussian process modeling with inequality constraints. {\it Ann. Fac. Sci. Toulouse}, 21(2):529-555,2012. · Zbl 1279.60047
[21] M. Deisenroth and J. Ng. Distributed Gaussian processes. {\it Proceedings of the 32nd International Conference on Machine}{\it Learning, Lille, France. JMLR: W&CP volume 37}, 2015.
[22] R. Furrer, M. G. Genton, and D. Nychka. Covariance tapering for interpolation of large spatial datasets. {\it Journal of Compu-}{\it tational and Graphical Statistics}, 2006.
[23] S. Golchi, D.R. Bingham, H. Chipman, and D.A. Campbell. Monotone Emulation of Computer Experiments. {\it SIAM/ASA}{\it Journal on Uncertainty Quantification}, 3(1):370-392, 2015. · Zbl 1327.62146
[24] T. Hangelbroek, F. J. Narcowich, and J. D. Ward. Kernel approximation on manifolds I: bounding the lebesgue constant.{\it SIAM J. Math. Anal.}, 42:1732-1760, 2010. · Zbl 1219.41003
[25] P. Hennig and C. J. Schuler. Entropy search for information-efficient global optimization. {\it Journal of Machine Learning Re-}{\it search}, 13:1809-1837, 2012. · Zbl 1432.65073
[26] J. Hensman, N. Fusi, and N. D Lawrence. Gaussian processes for big data. {\it Uncertainty in Artificial Intelligence}, pages 282-290,2013.
[27] G. E. Hinton. Training products of experts by minimizing contrastive divergence. {\it Neural computation}, 14(8):1771-1800, 2002. · Zbl 1010.68111
[28] C. Jidling, N. Wahlström, A. Wills, and T B. Schön. Linearly constrained Gaussian processes. {\it arXiv preprint arXiv:1703.00787},2017.
[29] D.R. Jones, M. Schonlau, and W.J. Welch. Efficient global optimization of expensive black box functions. {\it Journal of Global}{\it Optimization}, 13:455-492, 1998. · Zbl 0917.90270
[30] C. G. Kaufman, D. Bingham, S. Habib, K. Heitmann, and J. A. Frieman. Efficient emulators of computer experiments usingcompactly supported correlation functions, with an application to cosmology. {\it The Annals of Applied Statistics}, 5(4):2470-2492,2011. · Zbl 1234.62166
[31] C. G. Kaufman, M. J. Schervish, and D. W. Nychka. Covariance tapering for likelihood-based estimation in large spatial datasets. {\it Journal of the American Statistical Association}, 103(484):1545-1555, 2008. · Zbl 1286.62072
[32] W-L. Loh. Fixed-domain asymptotics for a subclass of Matérn-type Gaussian random fields. {\it The Annals of Statistics}, 33:2344-2394, 2005. · Zbl 1086.62111
[33] W-L. Loh and T-K. Lam. Estimating structured correlation matrices in smooth Gaussian random field models. {\it The Annals of}{\it Statistics}, 28:880-904, 2000. · Zbl 1105.62376
[34] H. Maatouk and X. Bay. {\it A New Rejection Sampling Method for Truncated Multivariate Gaussian Random Variables Restricted}{\it to Convex Sets}, volume 163, pages 521-530. In: Cools R and Nuyens R (Eds). Springer International Publishing, Cham, 2016. · Zbl 1356.65012
[35] H. Maatouk and X. Bay. Gaussian process emulators for computer experiments with inequality constraints. {\it Mathematical}{\it Geosciences}, 49(5):1-26, 2017. · Zbl 1371.65006
[36] H. Maatouk and Y. Richet. constrKriging. {\it R package available online at https://github.com/maatouk/constrKriging}, 2015.
[37] H. Maatouk, O. Roustant, and Y. Richet. Cross-Validation Estimations of Hyper-Parameters of Gaussian Processes withInequality Constraints. {\it Procedia Environmental Sciences}, 27:38 - 44, 2015. Spatial Statistics conference 2015.
[38] K. V. Mardia and R. J. Marshall. Maximum likelihood estimation of models for residual covariance in spatial regression.{\it Biometrika}, 71:135-146, 1984. · Zbl 0542.62079
[39] G. Matheron. {\it La Théorie des Variables Régionalisées et ses Applications}. Fasicule 5 in Les Cahiers du Centre de MorphologieMathématique de Fontainebleau. Ecole Nationale Supérieure des Mines de Paris, 1970.
[40] J. B. Močkus. {\it Bayesian approach to global optimization}. Mathematics and its applications. Kluwer Academic, 1989. · Zbl 0693.49001
[41] M. Morris. Gaussian surrogates for computer models with time-varying inputs and ouptputs. {\it Technometrics}, 54:42-50, 2012.
[42] C.R. Nelson and A.F. Siegel. Parsimonious modeling of yield curves. {\it The Journal of Business}, 60(4):pp. 473-489, 1987.
[43] T. Nickson, T. Gunter, C. Lloyd, M. A. Osborne, and S. Roberts. Blitzkriging: Kronecker-structured stochastic Gaussianprocesses. {\it arXiv preprint arXiv:1510.07965}, 2015.
[44] E. Parzen. {\it Stochastic processes}. Holden-Day series in probability and statistics. Holden-Day, San Francisco, London, Amster-dam, 1962. · Zbl 0107.12301
[45] R. Paulo, G. Garcia-Donato, and J. Palomo. Calibration of computer models with multivariate output. {\it Computational Statistics}{\it and Data Analysis}, 56:3959-3974, 2012. · Zbl 1255.62187
[46] C.E. Rasmussen and C.K.I. Williams. {\it Gaussian Processes for Machine Learning}. The MIT Press, Cambridge, 2006. · Zbl 1177.68165
[47] J. Riihimaki and A. Vehtari. Gaussian processes with monotonicity information. In {\it AISTATS}, volume 9 of {\it JMLR Proceedings},pages 645-652. JMLR.org, 2010.
[48] H. Rue and L. Held. {\it Gaussian Markov random fields, Theory and applications}. Chapman & Hall, 2005.ESAIM: PROCEEDINGS AND SURVEYS179 · Zbl 1093.60003
[49] D. Rullière, N. Durrande, F. Bachoc, and C. Chevalier. Nested Kriging estimations for datasets with large number of observa-tions. {\it Statistics and Computing}, 2017. · Zbl 1384.62246
[50] J. Sacks, W.J. Welch, T.J. Mitchell, and H.P. Wynn. Design and analysis of computer experiments. {\it Statistical Science}, 4:409-423, 1989. · Zbl 0955.62619
[51] T.J. Santner, B.J. Williams, and W.I. Notz. {\it The Design and Analysis of Computer Experiments}. Springer, New York, 2003. · Zbl 1041.62068
[52] N. Srinivas, A. Krause, S. Kakade, and M. Seeger. Information-theoretic regret bounds for Gaussian process optimization itsetting. {\it IEEE Transactions on Information Theory}, 58(5):3250-3265, 2012. · Zbl 1365.94131
[53] M. L. Stein. Statistical properties of covariance tapers. {\it Journal of Computational and Graphical Statistics}, 22(4):866-885,2013.
[54] M. L. Stein. Limitations on low rank approximations for covariance matrices of spatial data. {\it Spatial Statistics}, 8:1-19, 2014.
[55] M. L. Stein, J. Chen, and M. Anitescu. Stochastic approximation of score functions for Gaussian processes. {\it The Annals of}{\it Applied Statistics}, 7(2):1162-1191, 2013. · Zbl 1454.62283
[56] M.L. Stein. {\it Interpolation of Spatial Data: Some Theory for Kriging}. Springer, New York, 1999. · Zbl 0924.62100
[57] L. Svensson. Estimating and interpreting forward interest rates: Sweden 1992-1994. Technical report, National Bureau ofEconomic Research, 1994.
[58] V. Tresp. A bayesian committee machine. {\it Neural Computation}, 12(11):2719-2741, 2000.
[59] C. K. I. Williams and C. E. Rasmussen. {\it Gaussian Processes for Machine Learning}. MIT Press, 2006. · Zbl 1177.68165
[60] R. L. Winkler. The consensus of subjective probability distributions. {\it Management Science}, 15(2):B-61, 1968.
[61] R. L. Winkler. Combining probability distributions from dependent information sources. {\it Management Science}, 27(4):479-488,1981. · Zbl 0458.62007
[62] G. Xu and M. G. Genton. Tukey g-and-h random fields. {\it Journal of the American Statistical Association}, to appear, 2016.
[63] S. J. Yakowitz and F. Szidarovszky. A comparison of Kriging with nonparametric regression methods. {\it Journal of Multivariate}{\it Analysis}, 16:21-53, 1985. · Zbl 0591.62060
[64] Z. Ying. Asymptotic properties of a maximum likelihood estimator with data from a Gaussian process. {\it Journal of Multivariate}{\it Analysis}, 36:280-296, 1991. · Zbl 0733.62091
[65] Z. Ying. Maximum likelihood estimation of parameters under a spatial sampling scheme. {\it The Annals of Statistics}, 21:1567-1590, 1993. · Zbl 0797.62019
[66] H. Zhang and Y. Wang. Kriging and cross validation for massive spatial data. {\it Environmetrics}, 21:290-304, 2010.
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.