A DERIVATIVE-FREE ALGORITHM FOR UNCONSTRAINED OPTIMIZATION

In this paper a hybrid algorithm which combines the pattern search method and the genetic algorithm for unconstrained optimization is presented. The algorithm is a deterministic pattern search algorithm,but in the search step of pattern search algorithm,the trial points are produced by a way like the genetic algorithm. At each iterate, by reduplication,crossover and mutation, a finite set of points can be used. In theory,the algorithm is globally convergent. The most stir is the numerical results showing that it can find the global minimizer for some problems ,which other pattern search algorithms don＇t bear.     

对偶算法的进一步研究

The dual algorithm for minimax problems is further studied in this paper. The resulting theoretical analysis shows that the condition number of the corresponding Hessian of the smooth modified Lagrange function with changing parameter in the dual algorithm is proportional to the reciprocal of the parameter,which is very important for the efficiency of the dual algorithm. At last ,the numerical experiments are reported to validate the analysis results.     

T—S模糊控制器中各类算子对系统输出的影响

模糊控制系统的核心是模糊控制算法，对Takagi-Sugeno（以下简称T—S）模糊控制器，由于其合取算子与汇总算子选取的不同，使其算法区别很大。文中对几种常用T-S模糊控制算法的输出曲面及误差曲面进行比较，以确定这些算子变化对系统输出的影响。     

Monte Carlo approximation through Gibbs output in generalized linear mixed models

Geyer (J. Roy. Statist. Soc. 56 (1994) 291) proposed Monte Carlo method to approximate the whole likelihood function. His method is limited to choosing a proper reference point. We attempt to improve the method by assigning some prior information to the parameters and using the Gibbs output to evaluate the marginal likelihood and its derivatives through a Monte Carlo approximation. Vague priors are assigned to the parameters as well as the random effects within the Bayesian framework to represent a non-informative setting. Then the maximum likelihood estimates are obtained through the Newton Raphson method. Thus, out method serves as a bridge between Bayesian and classical approaches. The method is illustrated by analyzing the famous salamander mating data by generalized linear mixed models.     

剖面隐马氏模型训练算法的改进

利用剖面隐马氏模型获得多序列联配，一般需要经过初始化、训练、联配三个过程．然而，目前广泛采用的Baum—welch训练算法假设各条可观察序列互相独立，这与实际情况有所不符．本文对剖面隐马氏模型，给出可观察序列在互相不独立情况下的改进Baum—wlelch算法，在可观察序列两种特殊情况下（互相独立和一致依赖），得到了改进算法的具体表达式，讨论了一般情况下权重的选取方法．最后通过一个具体的蛋白质家族的多序列联配来说明改进算法的效果．     

线性规划的约束条件“滚雪球”算法

为使线性规划的每个约束条件部分或全部地拥有原整个约束条件所包含的信息,将线性规划的约束条件“滚雪球”后得到与原约束条件等价的新约束条件,对新约束条件所构成的线性规划采用目标函数最速递减算法.有一定规模的随机数值算例显示了该算法只需进行m(约束条件数)次迭代即可求得最优解.     

A NEW ALGORITHM OF RELAXATION METHOD FOR PARTICLE ANALYSIS FROM FORWARD SCATTERED LIGHT

1. Introduction In many cases, particle dimension and size distribution play an important role in application. The analysis of parti- cle size is therefore necessary for research and develop- ment and also for process quality control in powder tech- nology and for many particle systems (Wang et al., 2000). Many instruments have been developed and constructed for this purpose. With the rapid development and availabil- ity of laser techniques, optoelectronics and microcomputer science, light sc…     

A polynomial-time algorithm for the change-making problem

Optimally making change—representing a given value with the fewest coins from a set of denominations—is in general NP-hard. In most real money systems however, the greedy algorithm is optimal. We give a polynomial-time algorithm to determine, for a given coin system, whether the greedy algorithm is optimal.     

An acyclic days-off scheduling problem

This paper studies the days off scheduling problem when the demand for staffing fluctuates from day to another and when the number of total workdays is fixed in advance for each employee. The scheduling problem is then to allocate rests to employees with different days off policies: (1) two or three consecutive days off for each employee per week and (2) at least three consecutive days off for each employee per month. For each one, we propose a polynomial time algorithm to construct a solution if it exists. Received: April 2005 / Revised version: October 2005 AMS classification: 60K25, 60K30     

Componentwise adaptation for high dimensional MCMC

Summary  We introduce a new adaptive MCMC algorithm, based on the traditional single component Metropolis-Hastings algorithm and on our earlier adaptive Metropolis algorithm (AM). In the new algorithm the adaption is performed component by component. The chain is no more Markovian, but it remains ergodic. The algorithm is demonstrated to work well in varying test cases up to 1000 dimensions.