A decision-theoretic approach to screening and selection with common random numbers

We present Q-SASS, a quasi-Newton method for unconstrained stochastic optimization that does not rely on common random numbers. Most existing quasi-Newton approaches leverage common random numbers to construct second-order updates. However, motivated by challenges in variational quantum algorithms—where such coordination is not possible—we consider the setting in which function values and gradients are accessible only through noisy probabilistic zeroth- and first-order oracles, and no common random numbers can be exploited. We derive high-probability tail bounds on the iteration complexity of our algorithm for nonconvex, convex, and strongly convex (more generally, those satisfying the PL condition) objective functions. Finally, we demonstrate the empirical benefits of our quasi-Newton updating scheme on both synthetic and quantum chemistry problems.

We consider noisy optimization and some traditional variance reduction techniques aimed at improving the convergence rate, namely (i) common random numbers (CRN), which is relevant for population-based noisy optimization and (ii) stratified sampling, which is relevant for most noisy optimization problems. We present artificial models of noise for which common random numbers are very efficient, and artificial models of noise for which common random numbers are detrimental. We then experiment on a desperately expensive unit commitment problem. As expected, stratified sampling is never detrimental. Nonetheless, in practice, common random numbers nonetheless provided, by far, most of the improvement.

To analyze simulation runs which use the same random numbers, the blocking concept of experimental design is not needed. Instead, this paper applies a linear regression model with a nondiagonal covariance matrix. This covariance matrix does not need to have a specific pattern such as constant covariances. A simple example yields surprising results. The paper proposes a new framework for the error analysis. This framework consists of three factors (namely, common random numbers, replication, model validity), each with three levels.

Parametric sensitivity of biochemical networks is an indispensable tool for studying system robustness properties, estimating network parameters, and identifying targets for drug therapy. For discrete stochastic representations of biochemical networks where Monte Carlo methods are commonly used, sensitivity analysis can be particularly challenging, as accurate finite difference computations of sensitivity require a large number of simulations for both nominal and perturbed values of the parameters. In this paper we introduce the common random number (CRN) method in conjunction with Gillespie's stochastic simulation algorithm, which exploits positive correlations obtained by using CRNs for nominal and perturbed parameters. We also propose a new method called the common reaction path (CRP) method, which uses CRNs together with the random time change representation of discrete state Markov processes due to Kurtz to estimate the sensitivity via a finite difference approximation applied to coupled reaction paths that emerge naturally in this representation. While both methods reduce the variance of the estimator significantly compared to independent random number finite difference implementations, numerical evidence suggests that the CRP method achieves a greater variance reduction. We also provide some theoretical basis for the superior performance of CRP. The improved accuracy of these methods allows for much more efficient sensitivity estimation. In two example systems reported in this work, speedup factors greater than 300 and 10,000 are demonstrated.

We consider the problem of stochastic simulation optimization with common random numbers over a numerical search domain. We propose the Knowledge Gradient for Common Random Numbers (KG-CRN) sequential sampling algorithm, a simple elegant modification to the Knowledge Gradient that incorporates the use of correlated noise in simulation outputs with Gaussian Process meta-models. We compare this method against the standard Knowledge Gradient and a more recently proposed variation that allows for pairwise sampling. Our method significantly outperforms both baselines under identical laboratory conditions while greatly reducing computational cost compared to pairwise sampling.

Common random numbers in multivariate simulations

With the rapid development of computing technology, using parallel computing to solve large-scale ranking-and-selection (R&S) problems has emerged as an important research topic. However, direct implementation of traditionally fully sequential procedures in parallel computing environments may encounter various problems. First, the scheme of all-pairwise comparisons, which is commonly used in fully sequential procedures, requires a large amount of computation and significantly slows down the selection process. Second, traditional fully sequential procedures require frequent communication and coordination among processors, which are also not efficient in parallel computing environments. In this paper, we propose three modifications on one classical fully sequential procedure, Paulson’s procedure, to speed up its selection process in parallel computing environments. First, we show that if no common random numbers are used, then we can significantly reduce the computation spent on all-pairwise comparisons at each round. Second, by batching different alternatives, we show that we can reduce the communication cost among the processors, leading the procedure to achieve better performance. Third, to boost the procedure’s final-stage selection, when the number of surviving alternatives is less than the number of processors, we suggest to sample all surviving alternatives to the maximal number of observations that they should take. We show that, after these modifications, the procedure remains statistically valid and is more efficient compared with existing parallel procedures in the literature. Summary of Contribution: Ranking and selection (R&S) is a branch of simulation optimization, which is an important area of operations research. In recent years, using parallel computing to solve large-scale R&S problems has emerged as an important research topic, and this research topic is naturally situated in the intersection of computing and operations research. In this paper, we consider how to improve a fully sequential R&S procedure, namely, Paulson’s procedure, to reduce the high computational complexity of all-pairwise comparisons and the burden of frequent communications and coordination, so that the procedure is more suitable and more efficient in solving large-scale R&S problems using parallel computing environments that are becoming ubiquitous and accessible for ordinary users. The procedure designed in this paper appears more efficient than the ones available in the literature and is capable of solving R&S problems with over a million alternatives in a parallel computing environment with 96 processors. The paper also extended the theory of R&S by showing that the all-pairwise comparisons may be decomposed so that the computational complexity may be reduced significantly, which drastically improves the efficiency of all-pairwise comparisons as observed in numerical experiments.

ABSTRACTThe central composite design (CCD) is often used to estimate the second-order linear model. This paper uses a correlation induction strategy of common random numbers (CRN) in simulation experiment and utilizes the induced correlations to obtain better estimates for the second-order linear model. This strategy assigns the CRN to all design points in the CCD. An appropriate selection of the axial points in CCD makes the weighted least squares (WLS) estimator be equivalent to ordinary least squares (OLS) estimator in estimating the linear model parameters of CCD. We analytically investigate the efficiency of this strategy in estimation of model parameters. Under certain conditions, this correlation induction strategy yields better results than independent random number strategy in estimating model parameters except intercept. The simulation experiment on a selected model supports such results. We expect a suggested random number assignment is useful in application of CCD in simulation experiments.

Traditionally, exactness, numerical stability and speed are the three main criteria for evaluating algorithms for random variate generation. However, it is sometimes required that the algorithms provide correlation between generated variates for the purpose of inducing dependence among the output of simulation runs. The inverse transformation, which produces optimal correlation induction, often performs poorly in terms of the first three criteria. Algorithms based on composition, rejection, and special properties which often excel in terms of the first three criteria, tend to scramble the use of random numbers, causing many attempts at common random numbers, antithetic variates and external control variates to fail. The concept of obtaining correlation via algorithms other than the inverse transformation is examined here. To demonstrate feasibility, previously developed algorithms for Poisson and binomial random variate generation are modified to obtain both positive and negative correlation between runs....

Subset selection of best simulated systems

We consider optimization via simulation over a finite set of alternatives. We employ a Bayesian value-ofinformation approach in which we allow both correlated prior beliefs on the sampling means and correlated sampling. Correlation in the prior belief allow us to learn about an alternative’s value from samples of similar alternatives. Correlation in sampling, achieved through common random numbers, allows us to reduce the variance in comparing one alternative to another. We allow for a more general combination of both types of correlation than has been offered previously in the Bayesian ranking and selection literature. We do so by giving an exact expression for the value of information for sampling the difference between a pair of alternatives, and derive new knowledge-gradient methods based on this valuation. 1

This paper addresses discrete optimization via simulation. We show that allowing for both a correlated prior distribution on the means (e.g., with discrete Kriging models) and sampling correlation (e.g., with common random numbers, or CRN) can significantly improve the ability to quickly identify the best alternative. These two correlations are brought together for the first time in a highly sequential knowledge-gradient sampling algorithm, which chooses points to sample using a Bayesian value of information (VOI) criterion. We provide almost sure convergence guarantees as the number of samples grows without bound when parameters are known and provide approximations that allow practical implementation including a novel use of the VOI’s gradient rather than the response surface’s gradient. We demonstrate that CRN leads to improved optimization performance for VOI-based algorithms in sequential sampling environments with a combinatorial number of alternatives and costly samples.

Bayesian estimation of dynamic asset pricing models with informative observations

A widely applied method of experimental design in stochastic simulation is to use the same pseudo random-number stream for each of the systems or alternatives to be compared. If there are only two alternatives, there is a well-known simple method of statistical analysis. This paper presents Graybill's3 extension of that analysis technique to the case of several systems or alternatives and points out its applicability in the simulation environment. The technique is illustrated using a stochastic simula tion model of a small inventory system. An empirical comparison with other methods of design and analysis is included.

Gives an overview of the main techniques for improving the statistical efficiency of simulation estimators. Efficiency improvement is typically (but not always) achieved through variance reduction. We discuss methods such as common random numbers, antithetic variates, control variates, importance sampling, conditional Monte Carlo, stratified sampling, and some others, as well as the combination of certain of those methods. We also survey the recent literature on this topic.