Introduction to Stochastic Search and Optimization: Estimation, Simulation, and ControlJohn Wiley & Sons, 11. 3. 2005 - Počet stran: 618 * Unique in its survey of the range of topics. * Contains a strong, interdisciplinary format that will appeal to both students and researchers. * Features exercises and web links to software and data sets. |
Obsah
1 | |
2 Direct Methods for Stochastic Search | 34 |
3 Recursive Estimation for Linear Models | 65 |
4 Stochastic Approximation for Nonlinear RootFinding | 95 |
5 Stochastic Gradient Form of Stochastic Approximation | 126 |
6 Stochastic Approximation and the FiniteDifference Method | 150 |
7 Simultaneous Perturbation Stochastic Approximation | 176 |
8 AnnealingType Algorithms | 208 |
15 SimulationBased Optimization II Stochastic Gradient and Sample Path Methods | 409 |
16 Markov Chain Monte Carlo | 436 |
17 Optimal Design for Experimental Inputs | 464 |
Appendix A Selected Results from Multivariate Analysis | 505 |
Appendix B Some Basic Tests in Statistics | 515 |
Appendix C Probability Theory and Convergence | 526 |
Appendix D Random Number Generation | 538 |
Appendix E Markov Processes | 547 |
9 Evolutionary Computation I Genetic Algorithms | 231 |
10 Evolutionary Computation 11 General Methods and Theory | 259 |
11 Reinforcement Learning via Temporal Differences | 278 |
12 Statistical Methods for Optimization in Discrete Problems | 300 |
13 Model Selection and Statistical Information | 329 |
14 SimulationBased Optimization I Regeneration Common Random Numbers and Selection Methods | 367 |
Answers to Selected Exercises | 552 |
558 | |
Frequently Used Notation | 580 |
583 | |
Další vydání - Zobrazit všechny
Introduction to Stochastic Search and Optimization: Estimation, Simulation ... James C. Spall Náhled není k dispozici. - 2005 |
Běžně se vyskytující výrazy a sousloví
algorithm analysis Appendix applications approach asymptotic averaging basic batch Chapter chromosomes coefficients components computed Consider convergence covariance matrix criterion CRNs cross-validation D-optimal defined definition density deterministic difficult discussed in Section efficiency elements error evaluations example FDSA final find finding finite first fitness fixed gain sequence Gibbs sampling global gradient approximation gradient estimate Hence Hessian Hessian matrix implementation independent information matrix input iterations linear models loss function loss measurements loss values Markov Markov chain methods Monte Carlo multiple comparisons multivariate noise noise-free noisy measurements nonlinear normally distributed null hypothesis optimal design output P-value parameters perturbation precision matrix prediction probability provides random numbers random search random variables recursive reflect relative replications represents root-finding sample mean satisfied scalar search and optimization significant simulated annealing simulation solution specific SPSA statistical step stochastic approximation stochastic gradient stochastic search Subsection sufficient supervised learning Suppose Theorem theory variance
Oblíbené pasáže
Strana 561 - Adaptive identification and control algorithms for nonlinear bacterial growth systems.
Strana 6 - Because of the inherent limitations of the vast majority of optimization algorithms, it is usually only possible to ensure that an algorithm will approach a local minimum with a finite amount of resources being put into the optimization process. However, since the local minimum may still yield a significantly improved solution (relative to no formal optimization process at all), the local minimum may be a fully acceptable solution for the resources available (human time, money, computer time, etc.)...
Odkazy na tuto knihu
Probabilistic and Randomized Methods for Design under Uncertainty Giuseppe Calafiore Náhled není k dispozici. - 2006 |