Search results for "Monte Carlo integration"

A new strategy for effective learning in population Monte Carlo sampling

2016

In this work, we focus on advancing the theory and practice of a class of Monte Carlo methods, population Monte Carlo (PMC) sampling, for dealing with inference problems with static parameters. We devise a new method for efficient adaptive learning from past samples and weights to construct improved proposal functions. It is based on assuming that, at each iteration, there is an intermediate target and that this target is gradually getting closer to the true one. Computer simulations show and confirm the improvement of the proposed strategy compared to the traditional PMC method on a simple considered scenario.

DeepXS: fast approximation of MSSM electroweak cross sections at NLO

2018

We present a deep learning solution to the prediction of particle production cross sections over a complicated, high-dimensional parameter space. We demonstrate the applicability by providing state-of-the-art predictions for the production of charginos and neutralinos at the Large Hadron Collider (LHC) at the next-to-leading order in the phenomenological MSSM-19 and explicitly demonstrate the performance for $pp\to\tilde{\chi}^+_1\tilde{\chi}^-_1,$ $\tilde{\chi}^0_2\tilde{\chi}^0_2$ and $\tilde{\chi}^0_2\tilde{\chi}^\pm_1$ as a proof of concept which will be extended to all SUSY electroweak pairs. We obtain errors that are lower than the uncertainty from scale and parton distribution functi…

Simple sampling Monte Carlo methods

2005

Correlation of primary relaxations and high-frequency modes in supercooled liquids. I. Theoretical background of a nuclear magnetic resonance experim…

2006

The question regarding a possible correlation of the time scales of primary and secondary relaxations in supercooled liquids is formulated quantitatively. It is shown how this question can be answered using spin-lattice relaxation weighted stimulated-echo experiments, which are presented in an accompanying paper [A. Nowaczyk, B. Geil, G. Hinze, and R. Böhmer, Phys. Rev. E 74, 041505 (2006)]. General theoretical expressions relevant for the description of such experiments in the presence of correlation effects are derived. These expressions are analyzed by Monte Carlo integration for various correlation scenarios also including exchange processes, which are the hallmark of dynamical heteroge…

Rotational Motion of Linear Molecules in Three Dimensions. A Path-Integral Monte Carlo Approach

1994

Abstract A path-integral Monte Carlo (PIMC) simulation method for the rotational motion of linear molecules in three dimensions is presented. The technique is applied to an H2 impurity in a static crystal-field. The resulting orientational distributions from quantum and classical simulations are obtained and discussed. The algorithm suffers from the “sign problem” of quantum simulations. However, as can be seen by comparing the low temperature simulation result to the variational solution of the Schrodinger equation, the PIMC method captures the quantum fluctuations.

Gibbs-ensemble path-integral Monte Carlo simulations of a mixed quantum-classical fluid

1995

We study a model fluid with classical translational degrees of freedom and internal quantum states in two spatial dimensions. The path-integral Monte Carlo and the Gibbs-ensemble Monte Carlo techniques are combined to investigate the liquid-gas coexistence region in this mixed quantum-classical system. A comparison with the phase diagram obtained in the canonical ensemble is also presented.

HOW MONTE CARLO SIMULATIONS CAN CLARIFY COMPLEX PROBLEMS IN STATISTICAL PHYSICS

2001

Statistical mechanics of condensed matter systems in physics (fluids and solids) derives macroscopic equilibrium properties of these systems as averages computed from a Hamiltonian that describes the atomistic interactions in the system. While analytic methods for most problems involve uncontrolled approximations, Monte Carlo simulations allow numerically exact treatments, apart from statistical errors and from the systematic problem that finite systems are treated rather than the thermodynamic limit. However, this problem can be overcome by finite size scaling methods, and thus Monte Carlo methods have become a very powerful tool to study even complex phase transitions. Examples given wil…

More on importance sampling Monte Carlo methods for lattice systems

2009

Minimal mass size of a stable He-3 cluster

2005

The minimal number of 3He atoms required to form a bound cluster has been estimated by means of a Diffusion Monte Carlo procedure within the fixed-node approximation. Several importance sampling wave functions have been employed in order to consider different shell-model configurations. The resulting upper bound for the minimal number is 32 atoms.

Glueball masses from ratios of path integrals

2011

By generalizing our previous work on the parity symmetry, the partition function of a Yang-Mills theory is decomposed into a sum of path integrals each giving the contribution from multiplets of states with fixed quantum numbers associated to parity, charge conjugation, translations, rotations and central conjugations. Ratios of path integrals and correlation functions can then be computed with a multi-level Monte Carlo integration scheme whose numerical cost, at a fixed statistical precision and at asymptotically large times, increases power-like with the time extent of the lattice. The strategy is implemented for the SU(3) Yang-Mills theory, and a full-fledged computation of the mass and …