Abstract
Although importance sampling is an established and effective sampling and estimation technique, it becomes unstable and unreliable for high-dimensional problems. The main reason is that the likelihood ratio in the importance sampling estimator degenerates when the dimension of the problem becomes large. Various remedies to this problem have been suggested, including heuristics such as resampling. Even so, the consensus is that for large-dimensional problems, likelihood ratios (and hence importance sampling) should be avoided. In this paper we introduce a new adaptive simulation approach that does away with likelihood ratios, while retaining the multi-level approach of the cross-entropy method. Like the latter, the method can be used for rare-event probability estimation, optimization, and counting. Moreover, the method allows one to sample exactly from the target distribution rather than asymptotically as in Markov chain Monte Carlo. Numerical examples demonstrate the effectiveness of the method for a variety of applications.
Similar content being viewed by others
References
Botev ZI (2007a) Nonparametric density estimation via diffusion mixing. Technical Report, The University of Queensland (http://espace.library.uq.edu.au/view/UQ:120006)
Botev ZI (2007b) Three examples of a practical exact markov chain sampling. Preprint, The University of Queensland (http://espace.library.uq.edu.au/view/UQ:130865)
Botev ZI, Kroese DP, Taimre T (2007) Generalized cross-entropy methods for rare-event simulation and optimization. Simulation 83(11):785–806
Bucklew JA (2004) Introduction to rare event simulation. Springer, Berlin Heidelberg, New York
Devroye L (1986) Non-uniform random variate generation. Springer, Berlin Heidelberg, New York
Diaconis P, Holmes S (1995) Three examples of Monte Carlo Markov chains: at the interface between statistical computing, computer science, and statistical mechanics. In: Aldous et al (ed) Discrete probability and algorithms. Springer, Berlin Heidelberg, New York, pp 43–56
Doucet A, de Freitas N, Gordon N (2001) Sequential Monte Carlo methods in practice. Springer, Berlin Heidelberg, New York
Efron B, Tibshirani R (1994) An introduction to the bootstrap. Chapman & Hall, New York
Kahn H, Harris TE (1951) Estimation of particle transmission by random sampling. National Bureau of Standards applied mathematics series. National Bureau of Standards, Washington, DC
Kroese DP, Porotsky S, Rubinstein RY (2006) The cross-entropy method for continuous multi-extremal optimization. Methodol Comput Appl Probab 8:383–407
Lieber D, Rubinstein RY, Elmakis D (1997) Quick estimation of rare events in stochastic networks. IEEE Trans Reliability 46:254–265
Olafsson S (2004) Two-stage nested partitions method for stochastic optimization. Methodol Comput Appl Probab 6:5–28
Propp JG, Wilson DB (1996) Exact sampling with coupled Markov chains and applications to statistical mechanics. Random Struct Algorithms 1–2:223–252
Robert CP, Casella G (2004) Monte Carlo statistical methods, 2nd edn. Springer, Berlin Heidelberg, New York
Ross SM (2002) Simulation, 3rd edn. Academic, New York
Rubinstein RY (1999) The cross-entropy method for combinatorial and continuous optimization. Methodol Comput Appl Probab 2:127–190
Rubinstein RY (2005) The stochastic minimum cross-entropy method for combinatorial optimization and rare-event estimation. Methodol Comput Appl Probab 7:5–50
Rubinstein RY (2006) How many needles are in a haystack, or how to solve #P-complete counting problems fast. Methodol Comput Appl Probab 8(1):5–51
Rubinstein RY, Kroese DP (2004) The cross-entropy method: a unified approach to combinatorial optimization, Monte Carlo simulation and machine learning. Springer, Berlin Heidelberg, New York
Rubinstein RY, Kroese DP (2007) Simulation and the Monte Carlo method, 2nd edn. Wiley, New York
Villén-Altamirano M, Villén-Altamirano J (1991) RESTART: a method for accelerating rare event simulations. In: Cohen JW, Pack CD (eds) Proceedings of the 13th International Teletraffic Congress, Queueing, Performance and Control in ATM, pp 71–76, Copenhagen, 1991
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Botev, Z.I., Kroese, D.P. An Efficient Algorithm for Rare-event Probability Estimation, Combinatorial Optimization, and Counting. Methodol Comput Appl Probab 10, 471–505 (2008). https://doi.org/10.1007/s11009-008-9073-7
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11009-008-9073-7
Keywords
- Likelihood ratio degeneracy
- Kernel density estimation
- Importance sampling
- Exact sampling
- Rare-event probability estimation
- Combinatorial counting