A Guide to Exact Simulation

Knowledge of the Earth's interior relies on indirect information collected at or near the surface. Typically, data do not uniquely constrain the subsurface properties and are contaminated by noise, and therefore the solution of ill-posed inverse problems is required. Geophysicists have traditionally addressed such problems through deterministic approaches, seeking a single 'best-fitting' model defined by optimality criteria that reflect our understanding of the problem at hand. In recent decades, Bayesian approaches have become increasingly common, as they characterise the posterior probability distribution of the model conditioned on the observations, thereby quantifying uncertainty.In this context, Markov chain Monte Carlo (MCMC) methods have emerged as a crucial tool as they allow sampling from posterior distributions of arbitrary complexity. At the core of many MCMC algorithms lies the Metropolis-Hastings scheme. This combines a proposal distribution with a probabilistic acceptance criterion to construct a Markov chain that has the desired target distribution as its stationary distribution. The algorithm is versatile as it rests on mild technical conditions on the proposal, and is thus widely adopted across a broad range of geoscientific inference problems. Yet when the parameter space is large or the forward models are computationally expensive---both common scenarios in geophysical applications---it can become inefficient, resulting in poor chain mixing and slow convergence to the target (posterior) distribution. These challenges underscore the importance of effective proposal mechanisms.In this presentation, we introduce a novel approach to designing Metropolis-Hastings proposals based on adaptive transport maps. The framework is inspired by recent developments from the field of Applied Mathematics linking Bayesian inference with Optimal Transport theory. The idea is to find a monotone, nonlinear transformation to recast a (complex) target probability distribution into a (simpler) reference distribution that is more amenable to standard MCMC steps. Our key contribution is to parameterise these transformations using invertible neural networks, ensuring monotonicity while gaining the flexibility and expressiveness that neural architectures afford.The proposed method progresses iteratively. We begin with a standard sampling strategy (e.g., a random-walk Metropolis) to obtain initial draws from the target distribution. These samples inform the training of an invertible neural network that learns to map from the target to a simpler reference distribution, specifically a standard (multivariate) Gaussian. Subsequent proposals are then generated in the reference space, either as global independence moves or local perturbations, and are accepted or rejected following a suitably modified Metropolis-Hastings criterion. As more samples accumulate, the network's parameters are updated, improving overall sampling efficiency. Ultimately, the approach yields not only an ensemble of samples representative of the desired target distribution---just as in standard MCMC---but also a compact, learned representation of it in the network's weights.We illustrate the proposed paradigm both theoretically and through examples. Preliminary results indicate that transport-map-enhanced MCMC has the potential to significantly accelerate Bayesian sampling across a range of applications.

Article Free Access Share on Exact sampling and approximate counting techniques Author: Mark Huber School of Operations Research and Industrial Engineering, Cornell University, Ithaca, NY School of Operations Research and Industrial Engineering, Cornell University, Ithaca, NYView Profile Authors Info & Claims STOC '98: Proceedings of the thirtieth annual ACM symposium on Theory of computingMay 1998 Pages 31–40https://doi.org/10.1145/276698.276709Published:23 May 1998Publication History 25citation666DownloadsMetricsTotal Citations25Total Downloads666Last 12 Months66Last 6 weeks15 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteeReaderPDF

The conventional Markov chain Monte Carlo (MCMC) method is limited to the selected shape and size of proposal distribution and is not easy to start when the initial proposal distribution is far away from the target distribution. To overcome these drawbacks of the conventional MCMC method, two useful improvements in MCMC method, adaptive Metropolis (AM) algorithm and delayed rejection (DR) algorithm, are attempted to be combined. The AM algorithm aims at adapting the proposal distribution by using the generated estimators, and the DR algorithm aims at enhancing the efficiency of the improved MCMC method. Based on the improved MCMC method, a Bayesian amplitude versus offset (AVO) inversion method on the basis of the exact Zoeppritz equation has been developed, with which the P- and S-wave velocities and the density can be obtained directly, and the uncertainty of AVO inversion results has been estimated as well. The study based on the logging data and the seismic data demonstrates the feasibility and robustness of the method and shows that all three parameters are well retrieved. So the exact Zoeppritz-based nonlinear inversion method by using the improved MCMC is not only suitable for reservoirs with strong-contrast interfaces and long-offset ranges but also it is more stable, accurate and anti-noise.

Density-estimation likelihood-free inference (DELFI) has recently been\nproposed as an efficient method for simulation-based cosmological parameter\ninference. Compared to the standard likelihood-based Markov Chain Monte Carlo\n(MCMC) approach, DELFI has several advantages: it is highly parallelizable,\nthere is no need to assume a possibly incorrect functional form for the\nlikelihood and complicated effects (e.g the mask and detector systematics) are\neasier to handle with forward models. In light of this, we present two DELFI\npipelines to perform weak lensing parameter inference with lognormal\nrealizations of the tomographic shear field -- using the C_l summary statistic.\nThe first pipeline accounts for the non-Gaussianities of the shear field,\nintrinsic alignments and photometric-redshift error. We validate that it is\naccurate enough for Stage III experiments and estimate that O(1000) simulations\nare needed to perform inference on Stage IV data. By comparing the second DELFI\npipeline, which makes no assumption about the functional form of the\nlikelihood, with the standard MCMC approach, which assumes a Gaussian\nlikelihood, we test the impact of the Gaussian likelihood approximation in the\nMCMC analysis. We find it has a negligible impact on Stage IV parameter\nconstraints. Our pipeline is a step towards seamlessly propagating all\ndata-processing, instrumental, theoretical and astrophysical systematics\nthrough to the final parameter constraints.\n

Clustering is a fundamental problem in statistics and machine learning, whose solution is commonly computed by the Expectation-Maximization (EM) method, which finds a locally optimal solution for an objective function called log-likelihood. Since the surface of the log-likelihood function is non convex, a stochastic search with Markov Chain Monte Carlo (MCMC) methods can help escaping locally optimal solutions. In this article, we tackle two fundamental conflicting goals: Finding higher quality solutions and achieving faster convergence. With that motivation in mind, we introduce an efficient algorithm that combines elements of the EM and MCMC methods to find clustering solutions that are qualitatively better than those found by the standard EM method. Moreover, our hybrid algorithm allows tuning model parameters and understanding the uncertainty in their estimation. The main issue with MCMC methods is that they generally require a very large number of iterations to explore the posterior of each model parameter. Convergence is accelerated by several algorithmic improvements which include sufficient statistics, simplified model parameter priors, fixing covariance matrices and iterative sampling from small blocks of the data set. A brief experimental evaluation shows promising results.

Markov Chain Monte Carlo (MCMC) is widely applied to fit complex, practical models in various disciplines, such as engineering, physics, agriculture, etc. Indeed, when the target density $f$ is not available in closed form, a Markov chain is constructed with stationary density $f$ for forming Monte Carlo estimators of means with respect to $f$. Although a Harris ergodic Markov chain will converge to the target distribution as the Monte Carlo sample size tends to infinity, and the population means can be consistently estimated by the corresponding MCMC estimators, in practice, it is important to estimate the standard errors associated with these Monte Carlo estimates due to the availability of only a finite sample. Therefore, the geometric ergodicity of a Markov Chain is highly desired as it is the most standard method for guaranteeing a central limit theorem (CLT) for MCMC estimators and finding its standard errors. Among the different MCMC algorithms, Metropolis-Hastings (MH) algorithms are predominant. This thesis establishes conditions under which MH algorithms with a position-dependent proposal covariance matrix will or will not have the geometric convergence rate. Indeed, we provide these results for MH algorithms where the normal proposal density has a general mean function $c(x)$ and a covariance matrix $G(x)$ depending on the current position $x$ of the Markov chain. For modern variants of the popular Metropolis adjusted Langevin algorithms (MALA), like the manifold MALA (MMALA) that adapt to the geometry of the target distributions, the proposal covariance matrix changes in every iteration. Thus, we provide conditions for the geometric ergodicity of different variations of the Langevin algorithms, e.g., MMALA, position-dependent MALA (PMALA), and the pre-conditioned MALA (PCMALA). We also provide conditions guaranteeing geometric ergodicity of the pre-conditioned unadjusted Langevin algorithms (PCULA). Generalized linear mixed models (GLMMs), where the response variable is allowed to follow a non-Gaussian distribution, are popular statistical models, although their likelihood functions are not available in closed form. GLMMs can handle the over-dispersion often present in real-world data, and MCMC methods are generally used to fit frequentist and Bayesian GLMMs. Since the likelihood function for GLMMs is not available in closed form, Monte Carlo EM or Monte Carlo maximum likelihood methods are often used for making inferences from GLMMs. Simulation from the conditional (on the data) distribution of the random effects is, in turn, needed for implementing the Monte Carlo EM or Monte Carlo maximum likelihood methods. In this thesis, our general conditions for geometric ergodicity of variants of Langevin algorithms are also verified in the context of conditional simulation from the two most popular GLMMs, namely the binomial GLMM with the logit link and the Poisson GLMM with the log link. These results have important practical implications as users can now confidently fit these models by attaching asymptotically valid standard errors with their Markov chain-based Monte Carlo estimates. Spatial generalized linear mixed models (SGLMMs) are GLMMs where the random effects form a spatial process and are useful for modeling spatially correlated data. We compare the performance of different variants of MALA mentioned above as well as other popular MH algorithms like Random Walk Metropolis in the context of conditional simulation from SGLMMs. Next, we consider MCMC algorithms for full Bayesian analysis of GLMMs. Blocking and reparameterization are techniques used to improve the efficiency and convergence properties of MCMC algorithms. We combine these strategies with pre-conditioning to develop efficient Langevin algorithms for Bayesian GLMMs and SGLMMs. We illustrate our algorithms using extensive real and simulated data analyses.

Time series models are often used in hydrology and meteorology studies to model streamflows series in order to make forecasting and generate synthetic series which are inputs for the analysis of complex water resources systems. In thispaper we introduce a new modeling approach for hydrologic and meteorological time series assuming a continuous distribution for the data, where both the conditional mean and conditional varianceparameters are modeled. Bayesian methods using standard MCMC (Markov Chain Monte Carlo Methods) are used to simulate samples for the joint posterior distribution of interest. Two applications to real data sets illustrate the proposedmethodology, assuming that the observations come from a normal, a gamma or a beta distribution. A first example is given by a time series of monthly averages of natural streamflows, measured in the year period ranging from1931 to 2010 in Furnas hydroelectric dam, Brazil. A second example is given with a time series of 313 air humidity data measured in a weather station of Rio Claro, a Brazilian city located in southeastern of Brazil. These applications motivate us to introduce new classes of models to analyze hydrological and meteorological time series

Markov chain Monte Carlo sampling using a reservoir method

We introduce a general Monte Carlo method based on Nested Sampling (NS), for sampling complex probability distributions and estimating the normalising constant. The method uses one or more particles, which explore a mixture of nested probability distributions, each successive distribution occupying ∼e −1 times the enclosed prior mass of the previous distribution. While NS technically requires independent generation of particles, Markov Chain Monte Carlo (MCMC) exploration fits naturally into this technique. We illustrate the new method on a test problem and find that it can achieve four times the accuracy of classic MCMC-based Nested Sampling, for the same computational effort; equivalent to a factor of 16 speedup. An additional benefit is that more samples and a more accurate evidence value can be obtained simply by continuing the run for longer, as in standard MCMC.

Background: Rapid progression in medical and health sciences have caused survival studies, where some patients have long-term survival, especially for chronic diseases such as breast cancer. Cure models can be applicable to analyze such data. Objectives: The aim of this study was to determine the risk factors associated with breast cancer, using mixture cure fraction model. Methods: We studied data for 438 patients, who were referred to cancer research center in Shahid Beheshti University of Medical Sciences. The patients were visited and treated during 1992 to 2012 and followed-up until October 2014. The data were analyzed by mixture cure fraction model based on GMW (generalized modified Weibull) distribution and inferences were obtained with Bayesian approach, using standard MCMC (Markov Chain Monte Carlo) methods. All analyses were performed, using SPSS v20 and OpenBUGS software. The significant level was considered at 0.05. Results: During the follow-up period, 75 (17.12%) deaths occurred by breast cancer and the one-year overall survival rate was 98%. Covariates such as numbers of metastatic lymph nodes and histologic grade were statistically significant. Also, the cure fraction estimation was obtained 58%. Conclusions: When some patients have a long-term survival, cure models can be an interesting model to study survival and these models estimate parameters better than the traditional models such as cox model. In this paper, the mixture cure fraction model based on GMW was fitted for analysing survival times in patients with breast cancer.

Airport air traffic is one of the most important and hardest one among all the airport data forecasts. In this paper, the Markov Chain Monte Carlo (MCMC) method of applied theory of statistics has been introduced into the aviation sector, and the discussion on airport air traffic forecast has been conducted taking Shanghai airport as the application background. MCMC method is designed to solve the problem parameters expectations where some baysian inference can not be directly calculated. This paper conducts the iterative computation using WinBUGS, EViews is used to estimate the initial iteration parameter of the regression model, and it is on this basis that the regression equation under the MCMC method is obtained.

As alluded to in the chapter “Linkage Analysis of Qualitative Traits”, neither the Elston–Steward algorithm nor the Lander–Green approach is amenable to genetic data from large complex pedigrees and a large number of markers. In such cases, Monte Carlo estimation methods provide a viable alternative to the exact solutions. Two types of Monte Carlo methods have been developed for linkage analysis, haplotype inference, and other kinds of genetic analysis. They are Markov chain Monte Carlo (MCMC) methods and Monte Carlo methods that are based on independent samples. Approaches based on Markov chain Monte Carlo methods are more widely applicable; there is practically no limit on the size or complexity of the pedigrees, nor on the number of markers to be considered simultaneously. In this chapter, we will review the basic principles of MCMC methods for multipoint linkage analysis with extended pedigrees. Both simulations and application to data from the Framingham study will be used to compare and contrast three MCMC software packages: LOKI, MORGAN, and SIMWALK.

Polya tree Monte Carlo method

Traditional full-waveform inversion (FWI) methods only render a “best-fit” model that cannot account for uncertainties of the ill-posed inverse problem. Additionally, local optimization-based FWI methods cannot always converge to a geologically meaningful solution unless the inversion starts with an accurate background model. We seek the solution for FWI in the Bayesian inference framework to address those two issues. In Bayesian inference, the model space is directly probed by sampling methods such that we obtain a reliable uncertainty appraisal, determine optimal models, and avoid entrapment in a small local region of the model space. The solution of such a statistical inverse method is completely described by the posterior distribution, which quantifies the distributions for parameters and inversion uncertainties. To efficiently sample the posterior distribution, we introduce a sampling algorithm in which the proposal distribution is constructed by the local gradient and the diagonal approximate Hessian of the local log posterior. Our algorithm is called the gradient-based Markov chain Monte Carlo (GMCMC) method. The GMCMC FWI method can quantify inversion uncertainties with estimated posterior distribution given sufficiently long Markov chains. By directly sampling the posterior distribution, we obtain a global view of the model space. Theoretically speaking, statistical assessments do not depend on starting models. Our method is applied to the 2D Marmousi model with the frequency-domain FWI setting. Numerical results suggest that our method can be readily applied to 2D cases with affordable computational efforts.

We propose to perform statistical probabilistic model checking by using perfect simulation in order to verify steady-state and time unbounded until formulas over Markov chains. The model checking of probabilistic models by statistical methods has received increased attention in the last years since it provides an interesting alternative to numerical model checking which is poorly scalable with the increasing model size. In previous statistical model checking works, unbounded until formulas could not be efficiently verified, and steady-state formulas had not been considered due to the burn-in time problem to detect the steady-state. Perfect simulation is an extension of Markov Chain Monte Carlo (MCMC) methods that allows us to obtain exact steady-state samples of the underlying Markov chain, and thus it avoids the burn-in time problem to detect the steady-state. Therefore we suggest to verify time unbounded until and steady-state dependability properties for large Markov chains through statistical model checking by combining perfect simulation and statistical hypothesis testing.