This is the proceedings of the " 8
th
IMACS Seminar on Monte Carlo Methods" held from August 29 to September 2, 2011 in Borovets, Bulgaria, and organized by the Institute of Information and Communication Technologies of the Bulgarian Academy of Sciences in cooperation with the International Association for Mathematics and Computers in Simulation (IMACS). Included are 24 papers which cover all topics presented in the sessions of the seminar: stochastic computation and complexity of high dimensional problems, sensitivity analysis, high-performance computations for Monte Carlo applications, stochastic metaheuristics for optimization problems, sequential Monte Carlo methods for large-scale problems, semiconductor devices and nanostructures.
The history of the IMACS Seminar on Monte Carlo Methods goes back to April 1997 when the first MCM Seminar was organized in Brussels:
1
st
IMACS Seminar, 1997, Brussels, Belgium
2
nd
IMACS Seminar, 1999, Varna, Bulgaria
3
rd
IMACS Seminar, 2001, Salzburg, Austria
4
th
IMACS Seminar, 2003, Berlin, Germany
5
th
IMACS Seminar, 2005, Tallahassee, USA
6
th
IMACS Seminar, 2007, Reading, UK
7
th
IMACS Seminar, 2009, Brussels, Belgium
8
th
IMACS Seminar, 2011, Borovets, Bulgaria
Inhaltsverzeichnis
1;Preface;5 2;1 Improvement of Multi-population Genetic Algorithms Convergence Time;15 2.1;1.1 Introduction;15 2.2;1.2 Short Overview of MpGA Modifications;16 2.3;1.3 Parameter Identification of S. cerevisiae Fed-Batch Cultivation Using Different Kinds of MpGA;18 2.4;1.4 Analysis and Conclusions;21 3;2 Parallelization and Optimization of 4D Binary Mixture Monte Carlo Simulations Using Open MPI and CUDA;25 3.1;2.1 Introduction;25 3.2;2.2 The Metropolis Monte Carlo Method;26 3.3;2.3 Decomposition into Subdomains and the Virtual Topology Using OpenMPI;27 3.4;2.4 Management of Hypersphere Coordinate Migration Between Domains;28 3.4.1;2.4.1 Communication between the CPU and the GPU;29 3.5;2.5 Pseudorandom Number Generation;29 3.6;2.6 Results of Running the Modified Code;29 3.7;2.7 Conclusions;32 4;3 Efficient Implementation of the Heston Model Using GPGPU;35 4.1;3.1 Introduction;35 4.2;3.2 Our GPGPU-Based Algorithm for Option Pricing;37 4.3;3.3 Numerical Results;39 4.4;3.4 Conclusions and Future Work;41 5;4 On a Game-Method for Modeling with Intuitionistic Fuzzy Estimations. Part 2;43 5.1;4.1 Introduction;43 5.2;4.2 Short Remarks on the Game-Method for Modeling from Crisp Point of View;43 5.3;4.3 On the Game-Method for Modeling with Intuitionistic Fuzzy Estimations;45 5.4;4.4 Main Results;48 5.5;4.5 Conclusion;50 6;5 Generalized Nets, ACO Algorithms, and Genetic Algorithms;53 6.1;5.1 Introduction;53 6.2;5.2 ACO and GA;54 6.3;5.3 GN for Hybrid ACO-GA Algorithm;56 6.4;5.4 Conclusion;58 7;6 Bias Evaluation and Reduction for Sample-Path Optimization;61 7.1;6.1 Introduction;61 7.2;6.2 Problem Formulation;63 7.3;6.3 Taylor-Based Bias Correction;65 7.4;6.4 Impact on the Optimization Bias;66 7.5;6.5 Numerical Experiments;67 7.6;6.6 Conclusions;69 8;7 Monte Carlo Simulation of Electron Transport in Quantum Cascade Lasers;73 8.1;7.1 Introduction;73 8.2;7.2 QCL Transport Model;73 8.2.1;7.2.1 Pauli Master Equation;74 8.2.2;7.2.2 Calculation of Basis States;75 8.2.3;7.2.3 Monte Carlo
Solver;76 8.3;7.3 Results and Discussion;78 8.4;7.4 Conclusion;79 9;8 Markov Chain Monte Carlo Particle Algorithms for Discrete-Time Nonlinear Filtering;83 9.1;8.1 Introduction;83 9.2;8.2 General Particle Filtering Framework;84 9.3;8.3 High Dimensional Particle Schemes;85 9.3.1;8.3.1 Sequential MCMC Filtering;85 9.3.2;8.3.2 Efficient Sampling in High Dimensions;86 9.3.3;8.3.3 Setting Proposal and Steering Distributions;87 9.4;8.4 Illustrative Examples;87 9.5;8.5 Conclusions;90 10;9 Game-Method for Modeling and WRF-Fire Model Working Together;93 10.1;9.1 Introduction;93 10.2;9.2 Description of the Game-Method for Modeling;94 10.3;9.3 General Description of the Coupled Atmosphere Fire Modeling and WRF-Fire;95 10.4;9.4 Wind Simulation Approach;97 10.5;9.5 Conclusion;98 11;10 Wireless Sensor Network Layout;101 11.1;10.1 Introduction;101 11.2;10.2 Wireless Sensor Network Layout Problem;102 11.3;10.3 ACO for WSN Layout Problem;104 11.4;10.4 Experimental Results;106 11.5;10.5 Conclusion;107 12;11 A Two-Dimensional Lorentzian Distribution for an Atomic Force Microscopy Simulator;111 12.1;11.1 Introduction;111 12.2;11.2 Modeling Oxidation Kinetics;112 12.3;11.3 Development of the Lorentzian Model;114 12.3.1;11.3.1 Algorithm for the Gaussian Model;114 12.3.2;11.3.2 Development of the Lorentzian Model;115 12.4;11.4 Conclusion;117 13;12 Stratified Monte Carlo Integration;119 13.1;12.1 Introduction;119 13.2;12.2 Numerical Integration;120 13.3;12.3 Conclusion;126 14;13 Monte Carlo Simulation of Asymmetric Flow Field Flow Fractionation;129 14.1;13.1 Motivation;129 14.2;13.2 AFFFF;130 14.3;13.3 Mathematical Model and Numerical Algorithm;131 14.3.1;13.3.1 Mathematical Model;131 14.3.2;13.3.2 The MLMC Algorithm;132 14.4;13.4 Numerical Results;133 15;14 Convexization in Markov Chain Monte Carlo;139 15.1;14.1 Introduction;139 15.2;14.2 Auxiliary Functions;140 15.2.1;14.2.1 Definition of Auxiliary Functions;140 15.2.2;14.2.2 Optimization Process for Auxiliary Functions;140 15.2.3;14.2.
3 Auxiliary Functions for Convex Functions;142 15.2.4;14.2.4 Objective Function Which Is the Sum of Convex and Concave Functions;142 15.3;14.3 Stochastic Auxiliary Functions;143 15.3.1;14.3.1 Stochastic Convex Learning (Summary);143 15.3.2;14.3.2 Auxiliary Stochastic Functions;144 15.4;14.4 Metropolis-Hastings Auxiliary Algorithm;144 15.5;14.5 Numerical Experiments;145 15.6;14.6 Conclusion;146 16;15 Value Simulation of the Interacting Pair Number for Solution of the Monodisperse Coagulation Equation;149 16.1;15.1 Introduction;149 16.2;15.2 Value Simulation for Integral Equations;151 16.2.1;15.2.1 Value Simulation of the Time Interval Between Interactions;152 16.2.2;15.2.2 VSIPN to Estimate the Monomer Concentration Jh1;153 16.2.3;15.2.3 VSIPN to Estimate the Monomer and Dimer Concentration Jh12;154 16.3;15.3 Results of the Numerical Experiments;155 16.4;15.4 Conclusion;157 17;16 Parallelization of Algorithms for Solving a Three-Dimensional Sudoku Puzzle;159 17.1;16.1 Introduction;159 17.2;16.2 The Simulated Annealing Method;160 17.3;16.3 Successful Algorithms for Solving the Three-Dimensional Puzzle Using MPI;161 17.3.1;16.3.1 An Embarrassingly Parallel Algorithm;162 17.3.2;16.3.2 Distributed Simulated Annealing Using a Master/Worker Organization;163 17.4;16.4 Results;163 17.5;16.5 Conclusions;166 18;17 The Efficiency Study of Splitting and Branching in the Monte Carlo Method;169 18.1;17.1 Introduction;169 18.2;17.2 Randomized Branching;170 18.3;17.3 Splitting;173 19;18 On the Asymptotics of a Lower Bound for the Diaphony of Generalized van der Corput Sequences;177 19.1;18.1 Introduction and Main Result;177 19.2;18.2 Definitions and Previous Results;179 19.3;18.3 Proof of Theorem 18.1;180 20;19 Group Object Tracking with a Sequential Monte Carlo Method Based on a Parameterized Likelihood Function;185 20.1;19.1 Motivation;185 20.2;19.2 Group Object Tracking within the Sequential Monte Carlo Framework;186 20.3;19.3 Measurement Likelihood for Group Object Tracking;187
20.3.1;19.3.1 Introduction of the Notion of the Visible Surface;188 20.3.2;19.3.2 Parametrization of the Visible Surface;189 20.4;19.4 Performance Evaluation;189 20.5;19.5 Conclusions;191 21;20 The Template Design Problem: A Perspective with Metaheuristics;195 21.1;20.1 Introduction;195 21.2;20.2 The Template Design Problem;196 21.3;20.3 Solving the TDP under Deterministic Demand;197 21.3.1;20.3.1 Representation and Evaluation;197 21.3.2;20.3.2 Metaheuristic Approaches;199 21.4;20.4 Experimental Results;200 21.5;20.5 Conclusions and Future Work;204 22;21 A Comparison of Simulated Annealing and Genetic Algorithm Approaches for Cultivation Model Identification;207 22.1;21.1 Introduction;207 22.2;21.2 Genetic Algorithm;208 22.3;21.3 Simulated Annealing;209 22.4;21.4 E. coli MC4110 Fed-Batch Cultivation Process Model;210 22.5;21.5 Numerical Results and Discussion;211 22.6;21.6 Conclusion;212 23;22 Monte Carlo Investigations of Electron Decoherence due to Phonons;217 23.1;22.1 Introduction;217 23.2;22.2 The Algorithms;219 23.2.1;22.2.1 Algorithm A;220 23.2.2;22.2.2 Algorithm B;221 23.2.3;22.2.3 Algorithm C;221 24;23 Geometric Allocation Approach for the Transition Kernel of a Markov Chain;227 24.1;23.1 Introduction;227 24.2;23.2 Geometric Approach;228 24.2.1;23.2.1 Reversible Kernel;230 24.2.2;23.2.2 Irreversible Kernel;231 24.3;23.3 Benchmark Test;231 24.4;23.4 Conclusion;233 25;24 Exact Sampling for the Ising Model at All Temperatures;237 25.1;24.1 Introduction;237 25.2;24.2 The Ising Model;238 25.3;24.3 Exact Sampling;241 25.4;24.4 The Random Cluster Model;242 25.5;24.5 Exact Sampling for the Ising Model;244