1) learning operator
学习算子
1.
By application of optimized learning operator, the convergence speed of the objective function has been fasted and the oscillation problem of objective function has also been solved during the iteration.
在网络训练过程中采用改进的 BP算法 ,通过对学习算子的优化搜索 ,大大提高了网络的收敛速度 ,解决了 BP算法迭代过程中目标函数振荡问题 。
2.
This paper presents hybrid gcnetic algorithm with learning operator for the nonlincar programming problems, which has the feature of converging to the optimization and high converence speed.
本文针对一般形式的非线性规划问题,提出一种具有学习算子的混合遗传算法,该算法具有收敛速度快且能收敛于最优解等特点。
3.
A repulsion operator is designed to keep the diversity of particle swarm,and an efficient learning operator is defined to improve the a.
以旅行商问题为例,提出了一种离散粒子群优化算法,根据优化问题及离散量的特点,对粒子的位置、速度等量及其运算规则进行了重新定义,为抑制早熟停滞现象,为粒子和粒子群分别定义了个体多样性和微观多样性,算法中定义了排斥算子来保持粒子群的多样性,使用高效的学习算子来提高算法的局部求精能力,使算法在空间探索和局部求精间取得了很好的平衡,与领域中的其它典型算法进行了仿真比较,结果表明,离散粒子群优化算法具有很好的性能。
2) study operator
学习算子
1.
An improved genetic algorithm based on"approximative elite","study operator"and"chaos operator"has been presented in this paper according to the features of reactive power optimization in power system.
根据电力系统无功优化问题的特点,提出了一种基于近似最优个体、学习算子和浑沌算子的改进遗传算法。
2.
An improved genetic algorithm based on examples array,study operator and study probability,which can solve network re- configuration effectively,has been presented in this paper.
提出了一种基于“学习范例数组”、“学习算子”和“学习率”的改进遗传算法,可以有效地解决配电网络重构问题。
3.
To solve the global optimization problem of non-linear programming,an improved genetic algorithm with study operator is proposed in this paper.
为了解决非线性规划问题的全局优化问题,该文提出利用具有学习算子的遗传算法来进行求解。
3) self-learning operator
自学习算子
1.
A particle swarm optimization algorithm with self-learning operator is designed to tackle the traveling salesman problem.
针对旅行商问题,提出了一种带自学习算子的粒子群优化算法,根据旅行商问题及离散量运算的特点,对粒子的位置、速度等量及其运算规则进行了重新定义,为抑制早熟停滞现象,定义了变异速度来保持粒子群的多样性,使用自学习算子来提高算法的局部求精能力,使算法在空间探索和局部求精间取得了较好的平衡,与领域中的其它典型算法进行了仿真比较,结果表明,该算法具有良好的性能。
2.
In this paper,the CSA was improved in following sides:(1) self-learning operator of immune network was introduced to enhance local exploitation ability; (2) uniform design method was adopted to generate the initial antibodies; (3) separation method of objective and constraints was introduced to handle state-variable constraints.
为此,拟引入免疫网络自学习算子,均匀设计方法,以及目标与约束分离的处理机制,构建改进的克隆选择算法(ICSA),并将其用于状态变量带约束的间歇反应器和乙醇生物反应器的动态优化等实例,效果良好。
3.
To improve the efficiency of the Particle Swarm Optimization,self-learning operator is constructed.
本文利用粒子群优化算法求解VRP,为了提高求解效率,通过构造自学习算子、微粒的重新编码及运算规则的重新定义,使PSO算法能够处理离散问题,把微粒群算法应用于VRP问题的求解中,通过仿真证明了提出方法求解VRP问题的有效性和优越性。
4) learn operator based on cloud
云学习算子
5) LSM
学习子空间算法
1.
Recognition of A Limited Chinese Character Set Based on Improved CLAFIC LSM Algorithm;
基于改进型CLAFIC学习子空间算法的有限汉字集识别
6) Quantum group learning algorithm
量子群学习算法
补充资料:凹算子与凸算子
凹算子与凸算子
concave and convex operators
凹算子与凸算子「阴~皿d阴vex.耳阳.勿韶;.留叮.肠疽“‘.小啊j阅雌口叹甲司 半序空间中的非线性算子,类似于一个实变量的凹函数与凸函数. 一个Banach空间中的在某个锥K上是正的非线性算子A,称为凹的(concave)(更确切地,在K上u。凹的),如果 l)对任何的非零元x任K,下面的不等式成立: a(x)u。(Ax续斑x)u。,这里u。是K的某个固定的非零元,以x)与口(x)是正的纯量函数; 2)对每个使得 at(x)u。续x《月1(x)u。,al,月l>0,成立的x‘K,下面的关系成立二 A(tx))(l+,(x,t))tA(x),0
说明:补充资料仅用于学习参考,请勿用于其它任何用途。
参考词条