Controlled MCMC for Optimal Sampling
نویسندگان
چکیده
In this paper we develop an original and general framework for automatically optimizing the statistical properties of Markov chain Monte Carlo (MCMC) samples, which are typically used to evaluate complex integrals. The Metropolis-Hastings algorithm is the basic building block of classical MCMC methods and requires the choice of a proposal distribution, which usually belongs to a parametric family. The correlation properties together with the exploratory ability of the Markov chain heavily depend on the choice of the proposal distribution. By monitoring the simulated path, our approach allows us to learn “on the fly” the optimal parameters of the proposal distribution for several statistical criteria.
منابع مشابه
A Novel MCMC Based Receiver for Large-Scale Uplink Multiuser MIMO Systems
In this paper, we propose low complexity algorithms based on Markov chain Monte Carlo (MCMC) technique for signal detection and channel estimation on the uplink in large scale multiuser multiple input multiple output (MIMO) systems with tens to hundreds of antennas at the base station (BS) and similar number of uplink users. A BS receiver that employs a randomized sampling method (which makes a...
متن کاملOptimal Control of Malaria Model with Drug Resistance in Presence of Parameter Uncertainty
Plasmodium falciparum malaria has developed resistance to almost all available antimalarial drugs despite controlled antimalarial drug use. This study presents a mathematical model of spread of antimalarial drug resistance with control using treated bed nets, indoor residual spraying and treatment of symptomatic individuals. A common pitfall for such epidemiological models, however, is the abse...
متن کاملRELIABILITY–BASED DESIGN OPTIMIZATION OF CONCRETE GRAVITY DAMS USING SUBSET SIMULATION
The paper deals with the reliability–based design optimization (RBDO) of concrete gravity dams subjected to earthquake load using subset simulation. The optimization problem is formulated such that the optimal shape of concrete gravity dam described by a number of variables is found by minimizing the total cost of concrete gravity dam for the given target reliability. In order to achieve this p...
متن کاملParallel Markov Chain Monte Carlo via Spectral Clustering
As it has become common to use many computer cores in routine applications, finding good ways to parallelize popular algorithms has become increasingly important. In this paper, we present a parallelization scheme for Markov chain Monte Carlo (MCMC) methods based on spectral clustering of the underlying state space, generalizing earlier work on parallelization of MCMC methods by state space par...
متن کاملcomparison between Gauss-Newton and Markov-chain onte Carlo–based methods for inverting spectral nduced-polarization data for Cole-Cole parameters
We have developed a Bayesian model to invert spectral induced-polarization SIP data for Cole-Cole parameters using Markov-chain Monte Carlo MCMC sampling methods. We compared the performance of the MCMC-based stochastic method with an iterative Gauss-Newton-based deterministic method for Cole-Cole parameter estimation through inversion of synthetic and laboratory SIP data. The Gauss-Newton-base...
متن کامل