• Communications for Statistical Applications and Methods
• /
• 제10권1호
• /
• pp.203-210
• /
• 2003

In a standard Metropolis-type Monte Carlo simulation, the proposal distribution cannot be easily adapted to "local dynamics" of the target distribution. To overcome some of these difficulties, Duane et al. (1987) introduced the method of hybrid Monte Carlo(HMC) which combines the basic idea of molecular dynamics and the Metropolis acceptance-rejection rule to produce Monte Carlo samples from a given target distribution. In this paper, using the HMC within Gibbs sampler, an asymptotical estimate of the smoothing mean and a general solution to state space modeling in Bayesian framework is obtaineds obtained.

• 한국정보통신학회논문지
• /
• 제9권6호
• /
• pp.1341-1345
• /
• 2005

비선형이고 정규분포에 따르지 않는 state-space모형분석에서 순차적 몬테 칼로(SMC)는 유용한 도구 중의 하나이다. 모수와 시그럴을 동시에 추정하기 위해 Monte Carlo particle filters를 사용할 수가 있다. 그러나 SMC는 여러단계의 반복을 요구하는 특별한 particle filtering 기법을 필요로 하게 된다. 본 논문은 particle filtering과 순차적 hybrid Monte Carlo(SHMC)을 결합하는 방법을 제시하고자 한다. 실험을 위해 짱뚱어 자료를 사용하였다.

• Nuclear Engineering and Technology
• /
• 제44권2호
• /
• pp.151-160
• /
• 2012

The advantages for using Monte Carlo methods to analyze full-core reactor configurations include essentially exact representation of geometry and physical phenomena that are important for reactor analysis. But this substantial advantage comes at a substantial cost because of the computational burden, both in terms of memory demand and computational time. This paper focuses on the challenges facing full-core Monte Carlo for keff calculations and the prospects for Monte Carlo becoming a routine tool for reactor analysis.

• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2000년도 추계학술발표회 논문집
• /
• pp.135-140
• /
• 2000

베이지안 신경망 모형(Bayesian Neural Networks Models)에서 주어진 입력값(input)은 블랙 박스(Black-Box)와 같은 신경망 구조의 각 층(layer)을 거쳐서 출력값(output)으로 계산된다. 새로운 입력 데이터에 대한 예측값은 사후분포(posterior distribution)의 기대값(mean)에 의해 계산된다. 주어진 사전분포(prior distribution)와 학습데이터에 의한 가능도함수(likelihood functions)를 통해 계산되어진 사후분포는 매우 복잡한 구조를 갖게 됨으로서 기대값의 적분계산에 대한 어려움이 발생한다. 이때 확률적 추정에 의한 근사 방법인 몬테칼로 적분을 이용한다. 이러한 방법으로서 Hybrid Monte Carlo 알고리즘은 우수한 결과를 제공하여준다(Neal 1996). 본 논문에서는 Hybrid Monte Carlo 알고리즘과 기존에 많이 사용되고 있는 Gibbs sampling, Metropolis algorithm, 그리고 Slice Sampling등의 몬테칼로 방법들을 비교한다.

• E2M - 전기 전자와 첨단 소재
• /
• 제10권9호
• /
• pp.922-929
• /
• 1997

As the size of semiconductor devices shrinks in the horizontal as well as vertical dimension it is difficult to estimate the transport-velocity of electron because they drift in non-equilibrium with a few scattering. In this paper HYbrid Monte Carlo simulator which employs the drift-diffusion model for hole-transport and Monte Carlo model for electron-transport in order to reduce the simulation time and increase the accuracy as well has been developed and applied to analyze the electron-transport in InAlAs/InGaAs HBT which is attractive for an ultra high speed active device in high speed optical fiber transmission systems in terms of the velocity and energy distribution as well as cutoff frequency.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제20권4호
• /
• pp.277-293
• /
• 2016

We develop an efficient numerical method for pricing the Derivative Linked Securities (DLS). The payoff structure of the hybrid DLS consists with a standard 2-Star step-down type ELS and the range accrual product which depends on the number of days in the coupon period that the index stay within the pre-determined range. We assume that the 2-dimensional Geometric Brownian Motion (GBM) as the model of two equities and a no-arbitrage interest model (One-factor Hull and White interest rate model) as a model for the interest rate. In this study, we employ the Monte Carlo simulation method with the Compute Unified Device Architecture (CUDA) parallel computing as the General Purpose computing on Graphic Processing Unit (GPGPU) technology for fast and efficient numerical valuation of DLS. Comparing the Monte Carlo method with single CPU computation or MPI implementation, the result of Monte Carlo simulation with CUDA parallel computing produces higher performance.

• 한국전기전자재료학회논문지
• /
• 제11권7호
• /
• pp.571-580
• /
• 1998

The numerical model that can describe the ignition of pseudospark discharge using hybrid fluid-particle(Monte Carlo )method has been developed. This model consists of the fluid expression for transport of electrons and ions and Poisson's equation in the electric field. The fluid equation determines the spatiotemporal dependence of charged particle densities and the ionization source term is computed using the Monte carlo method. This model has been used to study the evolution of a discharge in Argon at 0.5 torr, with an applied voltage if 1kV. The evolution process of the discharge has been divided into four phases along the potential distribution : (1) Townsend discharge, (2) plasma formation, (3) onset of hollow cathode effect, (4) plasma expansion. From the numerical results, the physical mechanisms that lead to the rapid rise in current associated with the onset of pseudospark could be identified.

• 항공우주시스템공학회지
• /
• 제7권4호
• /
• pp.1-11
• /
• 2013

In this study, probabilistic reliability analysis was conducted for hybrid rocket performance using Monte-Carlo Simulation. For the accuracy, reliability analysis was performed with experimental data. To simplify the analysis process, the oxidizer was supplied with constant pressure, so that pressure variation with time can be eliminated. And time-space averaged regression rate model was used. The regression rate is obtained with a series of experiments. For reliability analysis of thrust, constant exponent of regression rate is assumed that has probabilistic character. So, the efficiency of characteristic velocity has also probabilistic values. As a results, probability distribution of the thrust is obtained by Monte-Carlo simulation using random samples of the input parameter and validated under the 95% confidence level.

• 정보처리학회논문지B
• /
• 제8B권3호
• /
• pp.231-236
• /
• 2001

일반적인 다층 신경망에서 가중치의 갱신 알고리즘으로 사용하는 오류 역전과 방식은 가중치 갱신 결과를 고정된(fixed) 한 개의 값으로 결정한다. 이는 여러 갱신의 가능성을 오직 한 개의 값으로 고정하기 때문에 다양한 가능성들을 모두 수용하지 못하는 면이 있다. 하지만 모든 가능성을 확률적 분포로 표현하는 갱신 알고리즘을 도입하면 이런 문제는 해결된다. 이러한 알고리즘을 사용한 베이지안 신경망 모형(Bayesian Neural Networks Models)은 주어진 입력값(Input)에 대해 블랙 박스(Black-Box)와같은 신경망 구조의 각 층(Layer)을 거친 출력값(Out put)을 계산한다. 이 때 주어진 입력 데이터에 대한 결과의 예측값은 사후분포(posterior distribution)의 기댓값(mean)에 의해 계산할 수 있다. 주어진 사전분포(prior distribution)와 학습데이터에 의한 우도함수(likelihood functions)에 의해 계산한 사후확률의 함수는 매우 복잡한 구조를 가짐으로 기댓값의 적분계산에 대한 어려움이 발생한다. 따라서 수치해석적인 방법보다는 확률적 추정에 의한 근사 방법인 몬테 칼로 시뮬레이션을 이용할 수 있다. 이러한 방법으로서 Hybrid Monte Carlo 알고리즘은 좋은 결과를 제공하여준다(Neal 1996). 본 논문에서는 Hybrid Monte Carlo 알고리즘을 적용한 신경망이 기존의 CHAID, CART 그리고 QUEST와 같은 여러 가지 분류 알고리즘에 비해서 우수한 결과를 제공하는 것을 나타내고 있다.

• Nuclear Engineering and Technology
• /
• 제53권4호
• /
• pp.1259-1265
• /
• 2021

In the present study, a large-area hybrid gamma imaging system was designed by adopting coded aperture imaging on the basis of a large-area Compton camera to achieve high imaging performance throughout a broad energy range (100-2000 keV). The system consisting of a tungsten coded aperture mask and monolithic NaI(Tl) scintillation detectors was designed through a series of Geant4 Monte Carlo radiation transport simulations, in consideration of both imaging sensitivity and imaging resolution. Then, the performance of the system was predicted by Geant4 Monte Carlo simulations for point sources under various conditions. Our simulation results show that the system provides very high imaging sensitivity (i.e., low values for minimum detectable activity, MDA), thus allowing for imaging of low-activity sources at distances impossible with coded aperture imaging or Compton imaging alone. In addition, the imaging resolution of the system was found to be high (i.e., around 6°) over the broad energy range of 59.5-1330 keV.