Location:
Search - optima
Search list
Description: 最优化方法中的非线性规划的powell方法的matlab实现，可支持多维的参数，自己修改即可。-optimization of the nonlinear programming method of Matlab Powell realized, support multidimensional parameter can change themselves.
Platform: |
Size: 1072 |
Author: 鹰回九天 |
Hits:
Description: 数字水印的实现工具optimark-digital watermarking tool for the realization of optimark
Platform: |
Size: 4758528 |
Author: 祁石 |
Hits:
Description: 最优化方法中的非线性规划的powell方法的matlab实现，可支持多维的参数，自己修改即可。-optimization of the nonlinear programming method of Matlab Powell realized, support multidimensional parameter can change themselves.
Platform: |
Size: 1024 |
Author: 鹰回九天 |
Hits:
Description: Alexandre Linhares于1998 提出了一种新的仿生计算方法,即捕食搜索算法。捕食搜索策略很好地协调了局部搜索和全局搜索之间的转换，已成功应用于组合优化领域的旅行商问题和超大规模集成电路设计问题。-The predatory search strategy consists of scanning the solution space in a straightforward manner, but, as each new optima is found, restricting the consequent search to a very small area, probing for a consecutive optimization within the boundaries of that area.
Platform: |
Size: 1204224 |
Author: abrahamlau |
Hits:
Description: 支持向量机可以对样本进行的分类，具有很好的泛化能及，并且可以解决小样本学习问题，在学习过程中避免出现局部最优解-Support vector machine classification of samples can have very good generalization can be and, and can solve the small sample learning problems, in the learning process to avoid local optima
Platform: |
Size: 26624 |
Author: luyuzheng |
Hits:
Description: 模拟退火fortran程序,可以根据优化函数设置参数-C Simulated annealing is a global optimization method that distinguishes
C between different local optima. Starting from an initial point, the
C algorithm takes a step and the function is evaluated. When minimizing a
C function, any downhill step is accepted and the process repeats from this
C new point. An uphill step may be accepted. Thus, it can escape from local
C optima. This uphill decision is made by the Metropolis criteria. As the
C optimization process proceeds, the length of the steps decline and the
C algorithm closes in on the global optimum. Since the algorithm makes very
C few assumptions regarding the function to be optimized, it is quite
C robust with respect to non-quadratic surfaces. The degree of robustness
C can be adjusted by the user. In fact, simulated annealing can be used as
C a local optimizer for difficult functions.
Platform: |
Size: 12288 |
Author: meibujun |
Hits:
Description: Enhanced Ant Colony Based Algorithm for
Platform: |
Size: 256000 |
Author: pangkaa |
Hits:
Description: 实用异形件优化排样系统的研究与开发
对于多种不规则零件的优化排样，综合利用冲裁件排样中的图形识别和图形分析处理技术和矩形件的排
样优化技术，在对它们作一些合理的改造后，有机地组合构造出多种不规则零件的5种排样算法。
-Shaped pieces of practical research and development of optimum layout system for a variety of irregular parts of the optimal layout, comprehensive utilization of optimal layout for blanking of the image recognition and graphics analysis and processing techniques and optimal layout of rectangular pieces of technology, they make some reasonable transformation, construct a variety of organic combination of irregular parts of the 5 nesting algorithms.
Platform: |
Size: 302080 |
Author: 天下冰霜 |
Hits:
Description: GA比起SA ，最大的优势在于对个初始解，而且存在杂交和变异，让SA具有非常强的跳出局部最优解的能力。而且简单通用，健壮性强。但是待定的参数很多，而且计算速度比较慢。选择，杂交，变异算子的选取也很关键。-GA than SA, the biggest advantage of an initial solution, and there is hybridization and mutation, so that SA has a very strong ability to jump out of local optima. And simple generic, robust and strong. However, many parameters to be determined, and the calculation speed is slower. Selection, hybridization, mutation operator selection is also critical.
Platform: |
Size: 89088 |
Author: JLH |
Hits:
Description: Ruta mas corta, algoritmo para la busqueda de la ruta mas optima
Platform: |
Size: 7168 |
Author: daryljorge |
Hits:
Description: 遗传算法对PID参数寻优实现倒立摆控制。-these programms fulfil the GA methods to find the global optima of the control parameters for a pid controller of an inverted pendulum.
Platform: |
Size: 9216 |
Author: yazada |
Hits:
Description: 结合多智能体的学习、协调策略及粒子群算法，提出了一种基于多智能体粒子群优化的配电网络重构方法。该方法采用粒子群算法的拓扑结构来构建多智能体的体系结构，在多智能体系统中，每一个粒子作为一个智能体，通过与邻域的智能体竞争、合作。能够更快、更精确地收敛到全局最优解。粒子的更新规则减少了算法不可行解的产生，提高了算法效率。实验结果表明，该方法具有很高的搜索效率和寻优性能。-Combining the study of multi-agent technology，coordinating strategies with P$O，a Multi-Agent Particle Swarm Optimization(MA-PSO)algorithm is presented to handle distribution network reconfiguration problem．It applies Von Neuman architecture
of Particle Swarm Optimization algorithm to the composition of multi-agent system．An agent in MA-PSO
represents a particle to PSO and a candidate solution to the optimization problem．In order to decrease fitness value quickly，agents compete and cooper-ate with their agent of neighboring area．Making use of these agent—agent interactions，MA—PSO realizes the purpose of
minimizing
the value of objective function．The rules of particle renovating reduce unfeasible solution in the process
of particle renovating，which raises the algorithm efficiency satty．The experiment results indicate the
prominent efficiency and significant global optima searching performance of MS—PSO．
Platform: |
Size: 515072 |
Author: yirufang |
Hits:
Description: 遗传算法及其育种:GA于20世纪60年代由美国Michigan大学J.H.Holland教授[1]首先提出。它可广泛应用于人工智能、机器学习、函数的优化、自动控制等领域。GA的突出特点是将问题的解空间间通过编码转换为GA的搜索空间，把问题的解转换为生物的个体，并借助生物的遗传和进化理论，对多个个体同时进行选择、交叉和变异操作。这样，可以较快地搜索到最优解。但是，遗传算法易陷入局部最优。搜索效率还不是
-Genetic Algorithm and Breeding: GA 1960s first proposed by the University of Michigan, USA JHHolland professor [1]. It can be widely used in artificial intelligence, machine learning, optimization, automatic control functions. The salient features of the GA is the solution space of the problem by transcoding the GA search space, the solution of the problem of biological individuals, and with the help of bio-genetic and evolutionary theory, multiple individual selection, crossover and variation operation. In this way, you can quickly search for the optimal solution. However, the genetic algorithm is easily trapped into local optima. Search efficiency is not
Platform: |
Size: 723968 |
Author: chodayy |
Hits:
Description: CVPR2012_oral
Weakly Supervised Structured Output Learning for Semantic Segmentation-We address the problem of weakly supervised semantic
segmentation. The training images are labeled only by the
classes they contain, not by their location in the image. On
test images instead, the method must predict a class label
for every pixel. Our goal is to enable segmentation algorithms
to use multiple visual cues in this weakly supervised
setting, analogous to what is achieved by fully supervised
methods. However, it is difficult to assess the relative usefulness
of different visual cues from weakly supervised training
data. We define a parametric family of structured models,
where each model weighs visual cues in a different way. We
propose a Maximum Expected Agreement model selection
principle that evaluates the quality of a model from the family
without looking at superpixel labels. Searching for the
best model is a hard optimization problem, which has no
analytic gradient and multiple local optima. We cast it as
a Bayesian optimization problem and propose an
Platform: |
Size: 2200576 |
Author: 费炳超 |
Hits:
Description: 二维最大熵法和二维最小交叉熵法是目前常用的两种阈值分割方法, 但在某些时候因为两种方法获取的阈
值过高或者过低, 使得分割失效。针对此问题, 提出了基于二维最大熵法和二维最小交叉熵法结合的图像分割方法。
首先, 对二维最小交叉熵公式进行转化 然后, 利用多目标规划理论将这两种方法有机结合使得到的阈值既满足二维
最大熵原则, 又满足二维最小交叉熵原则 最后, 利用二维直方图的特点推导出新型递推算法搜索最佳阈值并降低计
算复杂度。-The thresho ld ing method based 2-D max imum entropy and the one based on 2-D m inimum cross
entropy are used w ide ly in im age segmentation today, bu t in some app lications, they fa il to segment im ages
because of too h igh or too low thresho lds. Therefore, we proposed an im age thresho ld ing m ethod based on
the comb ination of 2-D max imum entropy and 2-D m inimum cross entropy. Firstly, the formula of the 2-D
m in imum cross entropy w as transfo rmed, then 2-D max imum entropy and 2-D m inimum cross entropy w ere
combined together usingmult-i objective prog ramm ing theory so that the optima l thresho ld value could satisfy
the threshold requirem ent of the bo th. A new recursive algo rithm w as in ferred using the features o f the 2-D
h istogram in order to search the best thresho ld vecto r and to reduce the compu ting complex ity
Platform: |
Size: 1525760 |
Author: rambolyb |
Hits:
Description: ：智能算法如粒子群算法已被应用于PID控制器的参数优化，以弥补传统优化方法容易产生振荡和较大超调量
的不足，但是粒子群算法存在易于早熟的缺点，在分析量子粒子群算法的基础上，提出了使用量子粒子群算法优化PID控
制器的参数。为了兼顾控制系统的各项性能指标，根据控制器的实际要求对各项指标进行加权作为算法的目标函数，对
PID控制器进行多目标寻优。通过2个传递函数实例，分别使用z—N、粒子群算法和量子粒子群算法进行了PID控制器
参数优化设计，并对结果进行了分析。-Abstract：Heuristics such as particle swarnl optimization is employed to enhance the capability of traditional techniques，
which is easy to produce surge and big overshoot，but PS0 may be trapped in the local optima of the objective and lead to poor performance．
This paper propesed the quantum-behaved particle 8wsl in optimization for the parameter optimization of PID controller．
A fitness function containing performance indexes Was defined and the algorithm Was used in multi-object optimization of PID controllers．
Two examples were given to illustrate the design procedure and exhibit the effectiveness of the proposed method via tomo
parison study with the existing Z—N and PSO approaches．
Platform: |
Size: 380928 |
Author: dhskja |
Hits:
Description: 提出了一种基于生境人工鱼群算法的多峰问题优化算法．该算法融合了模拟退火、小生境技术的思想，并加入了变异算子和自动生成合适小生境半径机制-Since it is difficult to find all the optima when artificial fish SWalTn algorithm(AFSA)is used in multimodal
optimization．a niche artificial fish swalTn algorithm(NAFSA)based on basic AFSA is proposed．NAFSA combines the
niche technique and the simulated annealing method wim AFsA
Platform: |
Size: 231424 |
Author: zhangyan |
Hits:
Description: Convergence analysis and performance of the extended artificial
physics optimization algorithm.artificial physics optimization (EAPO), a population-based,
stochastic, evolutionary algorithm (EA) for multidimensional search and optimization.
EAPO extends the physicomimetics-based Artificial Physics Optimization (APO) algorithm
by including each individual’s best fitness history. Including the history improves EAPO’s
search capability compared to APO. EAPO and APO invoke a gravitational metaphor in
which the force of gravity may be attractive or repulsive, the aggregate effect of which is
to move individuals toward local and global optima. A proof of convergence is presented
that reveals the conditions under which EAPO is guaranteed to converge
Platform: |
Size: 2048 |
Author: sina valizade |
Hits:
Description: 利用OTSU算法与遗传算法相结合时图像分割的具体步骤，并且发现遗传算法在优化大津图像分割算法时存在着因为早熟而陷入局部最优解不足的情况。-Using the OTSU algorithm is combined with genetic algorithm for image segmentation concrete steps, and found genetic algorithm optimization Otsu image segmentation algorithm to be trapped in there because of premature local optima insufficient.
Platform: |
Size: 2048 |
Author: 谢尔顿 |
Hits:
Description: Prueba1 FletcherReeves
Hernández Santiago José
Maestría en Ciencias de la Computación
Septiembre / 2011
1. Comenzar con un punto arbitrario
2. Calcular Gradiente de Fi
3. Si el Gradiente Fi es igual a 0(converge), termina
4. Si el Gradiente Fi es !=0 continuar
5. Encontrar dirección de búsqueda
Si= -GradienteFi= - Gradiente F(Xi)
6. Determinar la Longitud Optima del incremento lamda(i) en dirección Si
X(i+1)=X(i)+lamda(i)*S(i)= X(i)-lamda(i)*Gradiente F(Xi)
7. Hacer i=2
8. Obtener Gradiente Fi
9. Calcular Si= -GradFi + ( [abs(GradFi)^2]/[abs(GradF(i-1))^2] )*S(i-1)
10. Determinar la Longitud Optima del incremento lamda(i) en dirección Si
X(i+1)=X(i)+lamda(i)*S(i)= X(i)-lamda(i)*Gradiente F(Xi)
7. Verificar Optimalidad de X(i+1)
Si es optimo, detener
Si no es optimo hacer i=i+1 e ir al paso 8- Prueba1 FletcherReeves
Hernández Santiago José
Maestría en Ciencias de la Computación
Septiembre / 2011
1. Comenzar con un punto arbitrario
2. Calcular Gradiente de Fi
3. Si el Gradiente Fi es igual a 0(converge), termina
4. Si el Gradiente Fi es !=0 continuar
5. Encontrar dirección de búsqueda
Si= -GradienteFi= - Gradiente F(Xi)
6. Determinar la Longitud Optima del incremento lamda(i) en dirección Si
X(i+1)=X(i)+lamda(i)*S(i)= X(i)-lamda(i)*Gradiente F(Xi)
7. Hacer i=2
8. Obtener Gradiente Fi
9. Calcular Si= -GradFi + ( [abs(GradFi)^2]/[abs(GradF(i-1))^2] )*S(i-1)
10. Determinar la Longitud Optima del incremento lamda(i) en dirección Si
X(i+1)=X(i)+lamda(i)*S(i)= X(i)-lamda(i)*Gradiente F(Xi)
7. Verificar Optimalidad de X(i+1)
Si es optimo, detener
Si no es optimo hacer i=i+1 e ir al paso 8
Platform: |
Size: 6144 |
Author: JHSantiago |
Hits: