228,922 research outputs found
SMCTC : sequential Monte Carlo in C++
No full textSequential Monte Carlo methods are a very general class of Monte Carlo methods for sampling from sequences of distributions. Simple examples of these algorithms are used very widely in the tracking and signal processing literature. Recent developments illustrate that these techniques have much more general applicability, and can be applied very effectively to statistical inference problems. Unfortunately, these methods are often perceived as being computationally expensive and difficult to implement. This article seeks to address both of these problems. A C++ template class library for the efficient and convenient implementation of very general Sequential Monte Carlo algorithms is presented. Two example applications are provided: a simple particle filter for illustrative purposes and a state-of-the-art algorithm for rare event estimation
SMCTC: Sequential Monte Carlo in C++
No full textSequential Monte Carlo methods are a very general class of Monte Carlo methods for sampling from sequences of distributions. Simple examples of these algorithms are used very widely in the tracking and signal processing literature. Recent developments illustrate that these techniques have much more general applicability, and can be applied very effectively to statistical inference problems. Unfortunately, these methods are often perceived as being computationally expensive and difficult to implement. This article seeks to address both of these problems. A C++ template class library for the efficient and convenient implementation of very general Sequential Monte Carlo algorithms is presented. Two example applications are provided: a simple particle filter for illustrative purposes and a state-of-the-art algorithm for rare event estimation.
Speculative moves : multithreading Markov Chain Monte Carlo programs
No full textThe increasing availability of multi-core and multi-processor architectures provides new opportunities for improving the performance of many computer simulations. Markov Chain Monte Carlo (MCMC) simulations are widely used for approximate counting problems, Bayesian inference and as a means for estimating very high-dimensional integrals. As such MCMC has found a wide variety of applications in biology and biomedical analysis.
This paper presents a new method for reducing the runtime of Markov Chain Monte Carlo simulations by using SMP machines to speculatively perform iterations in parallel, reducing the runtime of MCMC programs whilst producing statistically identical results to conventional sequential implementations. We calculate the theoretical reduction in runtime that may be achieved using our technique under perfect conditions, and test and compare the method on a selection of multi-core and multi-processor architectures. Experiments are presented that show reductions in runtime of 35% using two cores and 55% using four cores
Oral History Interview with Monte C. Williamson, August 24, 1974
No full textInterview with Army Air Forces veteran Monte C. Williamson. The interview includes Williamson's personal experiences as a bombardier/navigator at Hickam Field during the Japanese attack at Pearl Harbor on December 7, 1941
Recommended from our members
Oral History Interview with Monte C. Williamson, August 24, 1974
No full textInterview with Army Air Forces veteran Monte C. Williamson. The interview includes Williamson's personal experiences as a bombardier/navigator at Hickam Field during the Japanese attack at Pearl Harbor on December 7, 1941
Why Monte Carlo Simulations are Inferences and not Experiments
No full textMonte Carlo Simulations arrive at their results by introducing randomness, sometimes derived from a physical randomizing device. Nonetheless, we argue, they open no new epistemic channels beyond that already employed by traditional simulations: the inference by ordinary argumentation of conclusions from assumptions built into the simulations. We show that Monte Carlo simulations cannot produce knowledge other than by inference; and that they resemble other computer simulations in the manner in which they derive their conclusions. Simple examples of Monte Carlo simulations are analyzed to identify the underlying inferences
Geodesic Monte Carlo on Embedded Manifolds
No full textMarkov chain Monte Carlo methods explicitly defined on the manifold of probability distributions have recently been established. These methods are constructed from diffusions across the manifold and the solution of the equations describing geodesic flows in the Hamilton–Jacobi representation. This paper takes the differential geometric basis of Markov chain Monte Carlo further by considering methods to simulate from probability distributions that themselves are defined on a manifold, with common examples being classes of distributions describing directional statistics. Proposal mechanisms are developed based on the geodesic flows over the manifolds of support for the distributions, and illustrative examples are provided for the hypersphere and Stiefel manifold of orthonormal matrices
Study on the effects of cooling phase and construction technology on the fire performance of R/C tunnels
No full textFire performance of tunnels can represent a critical issue in the design phase, even more than for other structures and infrastructures, due to some inherent features such as (a) the fire compartment geometry often leading to very high temperature (also making difficult the intervention of fire brigades), (b) the structural redundancy caused by soil restraint fostering the development of relevant indirect actions, and (c) the high compression state in the lining (all the more during the fire exposure) that increases the spalling propensity and severity. Within this context, the role played by key parameters such as lining thickness and stiffness are investigated by comparing the fire performance of two different technological solutions for the lining: (1) traditional cast-in-situ lining and (2) pre-cast segmental tunnel lining. The fire scenario also considers the cooling phase, in order to discuss the main critical points to be solved when facing the final stage of the fire. 3D finite element analyses have been performed, proving that (I) higher thickness and stiffness does not necessarily correspond to a higher safety factor due to the indirect actions, and (II) fire cooling phase (if any) can be even more critical than the heating phase
Monte Carlo Techniques in Studying Robust Estimators
No full textRecent work on robust estimation has led to many procedures, which are easy to formulate and straightforward to program but difficult to study analytically. In such circumstances experimental sampling is quite attractive, but the variety and complexity of both estimators and sampling situations make effective Monte Carlo techniques essential. This discussion examines problems, techniques, and results and draws on examples in studies of robust location and robust regression.
- …
