scholarly journals A Linear Regression Framework for the Verification of Bayesian Model Calibration Algorithms

2017 ◽  
Vol 2 (2) ◽  
Author(s):  
Jerry A. McMahan ◽  
Brian J. Williams ◽  
Ralph C. Smith ◽  
Nicholas Malaya
Keyword(s):  
Monte Carlo ◽  
Linear Regression ◽  
Bayesian Model ◽  
Model Calibration ◽  
Large Scale ◽  
Nonparametric Test ◽  
Observation Error ◽  
Large Scale Problems ◽  
Regression Framework ◽  
Reference Implementation

We describe a framework for the verification of Bayesian model calibration routines. The framework is based on linear regression and can be configured to verify calibration to data with a range of observation error characteristics. The framework is designed for efficient implementation and is suitable for verifying code intended for large-scale problems. We propose an approach for using the framework to verify Markov chain Monte Carlo (MCMC) software by combining it with a nonparametric test for distribution equality based on the energy statistic. Our matlab-based reference implementation of the framework is shown to correctly distinguish between output obtained from correctly and incorrectly implemented MCMC routines. Since correctness of output from an MCMC software depends on choosing settings appropriate for the problem-of-interest, our framework can potentially be used for verifying such settings.

LARGE SCALE MULTIKERNEL RELEVANCE VECTOR MACHINE FOR OBJECT DETECTION

2007 ◽  
Vol 16 (06) ◽  
pp. 967-979 ◽  
Author(s):  
DIMITRIS TZIKAS ◽  
ARISTIDIS LIKAS ◽  
NIKOLAS GALATSANOS
Keyword(s):  
Object Detection ◽  
Bayesian Model ◽  
Large Scale ◽  
Relevance Vector Machine ◽  
Fourier Domain ◽  
Inference Algorithm ◽  
Detection Problem ◽  
Large Scale Problems ◽  
Classification Tasks

The Relevance Vector Machine(RVM) is a widely accepted Bayesian model commonly used for regression and classification tasks. In this paper we propose a multikernel version of the RVM and present an alternative inference algorithm based on Fourier domain computation to solve this model for large scale problems, e.g. images. We then apply the proposed method to the object detection problem with promising results.

scholarly journals An optical observational cluster mass function at z ∼ 1 with the ORELSE survey

2021 ◽  
Vol 502 (3) ◽  
pp. 3942-3954
Author(s):  
D Hung ◽  
B C Lemaux ◽  
R R Gal ◽  
A R Tomczak ◽  
L M Lubin ◽  
...  
Keyword(s):  
Monte Carlo ◽  
Large Scale ◽  
Near Infrared ◽  
Function Analysis ◽  
Voronoi Tessellation ◽  
Cosmological Parameters ◽  
Mass Range ◽  
Mass Function ◽  
Cluster Mass ◽  
Theoretical Mass

ABSTRACT We present a new mass function of galaxy clusters and groups using optical/near-infrared (NIR) wavelength spectroscopic and photometric data from the Observations of Redshift Evolution in Large-Scale Environments (ORELSE) survey. At z ∼ 1, cluster mass function studies are rare regardless of wavelength and have never been attempted from an optical/NIR perspective. This work serves as a proof of concept that z ∼ 1 cluster mass functions are achievable without supplemental X-ray or Sunyaev-Zel’dovich data. Measurements of the cluster mass function provide important contraints on cosmological parameters and are complementary to other probes. With ORELSE, a new cluster finding technique based on Voronoi tessellation Monte Carlo (VMC) mapping, and rigorous purity and completeness testing, we have obtained ∼240 galaxy overdensity candidates in the redshift range 0.55 < z < 1.37 at a mass range of 13.6 < log (M/M⊙) < 14.8. This mass range is comparable to existing optical cluster mass function studies for the local universe. Our candidate numbers vary based on the choice of multiple input parameters related to detection and characterization in our cluster finding algorithm, which we incorporated into the mass function analysis through a Monte Carlo scheme. We find cosmological constraints on the matter density, Ωm, and the amplitude of fluctuations, σ8, of $\Omega _{m} = 0.250^{+0.104}_{-0.099}$ and $\sigma _{8} = 1.150^{+0.260}_{-0.163}$. While our Ωm value is close to concordance, our σ8 value is ∼2σ higher because of the inflated observed number densities compared to theoretical mass function models owing to how our survey targeted overdense regions. With Euclid and several other large, unbiased optical surveys on the horizon, VMC mapping will enable optical/NIR cluster cosmology at redshifts much higher than what has been possible before.

scholarly journals Identification of Efficient Sampling Techniques for Probabilistic Voltage Stability Analysis of Renewable-Rich Power Systems

Energies ◽  
2021 ◽  
Vol 14 (8) ◽  
pp. 2328
Author(s):  
Mohammed Alzubaidi ◽  
Kazi N. Hasan ◽  
Lasantha Meegahapola ◽  
Mir Toufikur Rahman
Keyword(s):  
Monte Carlo ◽  
Stability Analysis ◽  
Power Systems ◽  
Large Scale ◽  
Voltage Stability ◽  
Sampling Technique ◽  
Coefficient Of Determination ◽  
Sampling Techniques ◽  
Quasi Monte Carlo ◽  
Efficient Sampling

This paper presents a comparative analysis of six sampling techniques to identify an efficient and accurate sampling technique to be applied to probabilistic voltage stability assessment in large-scale power systems. In this study, six different sampling techniques are investigated and compared to each other in terms of their accuracy and efficiency, including Monte Carlo (MC), three versions of Quasi-Monte Carlo (QMC), i.e., Sobol, Halton, and Latin Hypercube, Markov Chain MC (MCMC), and importance sampling (IS) technique, to evaluate their suitability for application with probabilistic voltage stability analysis in large-scale uncertain power systems. The coefficient of determination (R2) and root mean square error (RMSE) are calculated to measure the accuracy and the efficiency of the sampling techniques compared to each other. All the six sampling techniques provide more than 99% accuracy by producing a large number of wind speed random samples (8760 samples). In terms of efficiency, on the other hand, the three versions of QMC are the most efficient sampling techniques, providing more than 96% accuracy with only a small number of generated samples (150 samples) compared to other techniques.

Sign in / Sign up

Export Citation Format

Share Document