Rare Event Probability Estimation in the Presence of Epistemic Uncertainty on Input Probability Distribution Parameters

详细信息    查看全文

• 作者：Mathieu Balesdent ; Jérôme Morio…
• 关键词：Epistemic uncertainty ; Surrogate model ; Importance sampling ; Rare event estimation ; Input–Output function ; Kriging
• 刊名：Methodology and Computing in Applied Probability
• 出版年：2016
• 出版时间：March 2016
• 年：2016
• 卷：18
• 期：1
• 页码：197-216
• 全文大小：788 KB
• 参考文献：Ackley DH (1987) A connectionist machine for genetic hillclimbing, vol 28. Kluwer Academic Publisher
  Arnold D (2002) Noisy optimization with evolution strategies. Springer
  Auger A, Finck S, Hansen N, Ros R (2010) Comparison tables: BBOB 2009 noisy testbed. INRIA technical report N∘384
  Balesdent M, Morio J, Marzat J (2013) Kriging-based adaptive importance sampling algorithms for rare event estimation. Struct Saf 44:1–10CrossRef
  Bucklew JA (2004) Introduction to rare event simulation. Springer
  Choi J, An D, Won J (2010) Bayesian approach for structural reliability analysis and optimization using the Kriging Dimension Reduction Method. J Mech Des 132:051003CrossRef
  Daniels HE (1954) Saddlepoint approximations in statistics. Ann Math Stat:631–650
  Davison A, Smith R (1990) Models for exceedances over high thresholds (with discussion). J R Stat Soc Ser B 52:393–442MathSciNet MATH
  De Boer PT, Kroese DP, Mannor S, Rubinstein RY (2005) A tutorial on the cross-entropy method. Ann Oper Res 134(1):19–67CrossRef MathSciNet MATH
  Dempster AP (1967) Upper and lower probabilities induced by a multivalued mapping. Ann Math Stat 38(2):325–339CrossRef MathSciNet MATH
  Deng G, Ferris M (2007) Extension of the direct optimization algorithm for noisy functions. In: Proceedings of the 2007 winter simulation conference. Winter Simulation Conference
  Der Kiureghian A, Ditlevsen O (2009) Aleatory or epistemic? Does it matter? Struct Saf 31(2):105–112CrossRef
  Ditlevsen O, Madsen HO (1996) Structural reliability methods. Wiley
  Dubourg V, Deheeger E, Sudret B (2011a) Metamodel-based importance sampling for the simulation of rare events. In: Faber, Kohler M J, Nishilima K (eds) Proceedings of the 11th international conference of statistics and probability in civil engineering (2011), Zurich, Switzerland
  Dubourg V, Sudret B, Bourinet J-M (2011b) Reliability-based design optimization using Kriging surrogates and subset simulation. Struct Multidiscip Optim 44(5):673–690CrossRef
  Ferson S, Oberkampf WL (2009) Validation of imprecise probability models. Int J Reliab Saf 3(1):3–22CrossRef
  Glasserman P, Heidelberger P, Shahabuddin P, Zajic T (1996) Splitting for rare event simulation: analysis of simple cases. In: Proceedings of the 1996 winter simulation conference. Winter Simulation Conference
  Glynn P (1996) Importance sampling for Monte Carlo estimation of quantiles. Publishing House of Saint Petersburg University
  Gomes H, Awruch A (2004) Comparison of response surface and neural network with other methods for structural reliability analysis. Struct Saf 26(1):49–67CrossRef
  Hansen N (2006) The CMA Evolution Strategy: a comparing review. In: Lozano J A, Larranaga P, Inza I, Bengoetxea E (eds) Towards a new evolutionary computation. Advances on estimation of distribution algorithms, pp 75–102
  Hansen N, Finck S, Ros R, Auger A (2009a) Real-parameter black-box optimization benchmarking 2009: noisy functions definitions. INRIA technical report N∘6869
  Hansen N, Niederberger A, Guzzella L, Koumoutsakos P (2009b) A method for handling uncertainty in evolutionary optimization with an application to feedback control of combustion. IEEE Trans Evol Comput 13(1):180–197CrossRef
  Helton JC (2011) Quantification of margins and uncertainties: conceptual and computational basis. Reliab Eng Syst Saf 96(9):976–1013CrossRef
  Helton JC, Johnson JD, Oberkampf WL (2004) An exploration of alternative approaches to the representation of uncertainty in model predictions. Reliab Eng Syst Saf 85(1):39–71CrossRef
  Janusevskis J, Le Riche R (2012) Simultaneous Kriging-based estimation and optimization of mean response. J Glob Optim 55(2):313–336CrossRef MathSciNet MATH
  Kruisselbrink JW, Reehuis E, Deutz A, Bäck T, Emmerich M (2011) Using the uncertainty handling cma-es for finding robust optima. In: Proceedings of the 13th annual conference on Genetic and evolutionary computation, GECCO ’11. ACM, New York
  Larson J (2012) Derivative-free optimization of noisy functions. Ph.D. thesis, University of Colorado
  L’Écuyer P, Demers V, Tuffin B (2006) Splitting for rare event simulation. In: Proceedings of the 2006 winter simulation conference. Winter Simulation Conference
  Li L, Bect J, Vazquez E (2010) A numerical comparison of two sequential Kriging-based algorithms to estimate a probability of failure. In: Uncertainty in computer model conference, Sheffield, UK, July, 12-14, 2010
  Limbourg P, De Rocquigny E, Andrianov G (2010) Accelerated uncertainty propagation in two-level probabilistic studies under monotony. Reliab Eng Syst Saf 95(9):998–1010CrossRef
  Lophaven S, Nielsen H, Songdergaard J (2002) DACE a MATLAB Kriging Toolbox. Technical Report IMM-TR-2002-12, Technical University of Denmark
  Lugannani R, Rice S (1980) Saddlepoint approximation for the distribution of the sum of independent random variables. Adv Appl Probab 12:475–490CrossRef MathSciNet MATH
  Matheron G (1963) Principles of geostatistics. Econ Geol 58(8):1246CrossRef
  Moore RE, Cloud MJ, Kearfott RB (2009) Introduction to interval analysis. SIAM
  Morio J (2010) How to approach the importance sampling density. Eur J Phys 31(2):41–48CrossRef MATH
  Morio J (2011) Non parametric adaptive importance sampling of the probability estimation of launcher impact position. Reliab Eng Syst Saf 96(1):178–183CrossRef
  Niederreiter H, Spanier J (2000) Monte Carlo and Quasi-Monte Carlo methods. Springer
  Picheny V (2009) Improving accuracy and compensating for uncertainty in surrogate modeling. Ph.D. thesis, University of Florida
  Pickands J (1975) Statistical inference using extreme order statistics. Ann Stat 3:119–131CrossRef MathSciNet MATH
  Rackwitz R (2001) Reliability analysis: a review and some perspectives. Struct Saf 23(4):365–395CrossRef
  Rajashekhar M, Ellingwood B (1993) A new look at the response surface approach for reliability analysis. Struct Saf 12(3):205–220CrossRef
  Rubinstein RY (1997) Optimization of computer simulation models with rare events. Eur J Oper Res 99(1):89–112CrossRef
  Rubinstein RY, Kroese DP (2004) The cross-entropy method : a unified approach to combinatorial optimization, Monte-Carlo simulation and machine learning (Information Science and Statistics). Springer
  Sasena M (2002) Flexibility and Efficiency enhancements for constrained global design optimization with Kriging approximation. Ph.D. thesis, University of Michigan
  Shafer G (1976) A mathematical theory of evidence, vol 1. Princeton University Press
  Sudret B (2012) Meta-models for structural reliability and uncertainty quantification. In: Asian-Pacific Symposium on Structural Reliability and its Applications, Singapore, Singapore 2012
  Thunnissen DP (2003) Uncertainty classification for the design and development of complex systems. In: 3rd annual predictive methods conference, Newport Beach, California, 2003
  Tomixk J, Arnold S, Barton R (1995) Sample size selection for improved Nelder-Mead performance. In: Proceedings of the 1995 winter simulation conference. Winter Simulation Conference
  Yao W, Chen X, Huang Y, Gurdal Z, van Tooren M (2013) Sequential optimization and mixed uncertainty analysis method for reliability-based optimization. AIAA J 51(9):2266–2277CrossRef
  Zadeh LA (1999) Fuzzy sets as a basis for a theory of possibility. Fuzzy Sets Syst 100:9–34CrossRef MathSciNet
  Zhang P (1996) Nonparametric importance sampling. J Am Stat Assoc 91(434):1245–1253CrossRef MathSciNet MATH
  
• 作者单位：Mathieu Balesdent (1)
  Jérôme Morio (3)
  Loïc Brevault (1) (2)
  
  1. Onera - The French Aerospace Lab, 91123, Palaiseau, France
  3. Onera - The French Aerospace Lab, BP 74025, 31055, Toulouse, France
  2. CNES - Launchers Directorate, Paris, France
  
• 刊物主题：Statistics, general; Life Sciences, general; Electrical Engineering; Economics general; Business/Management Science, general;
• 出版者：Springer US
• ISSN：1573-7713

文摘

The accurate estimation of rare event probabilities is a crucial problem in engineering to characterize the reliability of complex systems. Several methods such as Importance Sampling or Importance Splitting have been proposed to perform the estimation of such events more accurately (i.e., with a lower variance) than crude Monte Carlo method. However, these methods assume that the probability distributions of the input variables are exactly defined (e.g., mean and covariance matrix perfectly known if the input variables are defined through Gaussian laws) and are not able to determine the impact of a change in the input distribution parameters on the probability of interest. The problem considered in this paper is the propagation of the input distribution parameter uncertainty defined by intervals to the rare event probability. This problem induces intricate optimization and numerous probability estimations in order to determine the upper and lower bounds of the probability estimate. The calculation of these bounds is often numerically intractable for rare event probability (say 10−5), due to the high computational cost required. A new methodology is proposed to solve this problem with a reduced simulation budget, using the adaptive Importance Sampling. To this end, a method for estimating the Importance Sampling optimal auxiliary distribution is proposed, based on preceding Importance Sampling estimations. Furthermore, a Kriging-based adaptive Importance Sampling is used in order to minimize the number of evaluations of the computationally expensive simulation code. To determine the bounds of the probability estimate, an evolutionary algorithm is employed. This algorithm has been selected to deal with noisy problems since the Importance Sampling probability estimate is a random variable. The efficiency of the proposed approach, in terms of accuracy of the found results and computational cost, is assessed on academic and engineering test cases.