人工神经网络BP算法的改进和结构的自调整

本文解决了BP神经网络结构参数和学习速率的选取问题,并对传统的BP算法进行了改进,提出了BP神经网络动态全参数自调整学习算法,又将其编制成计算机程序,使得隐层节点和学习速率的选取全部动态实现,减少了人为因素的干预,改善了学习速率和网络的适应能力。计算结果表明：BP神经网络动态全参数自调整算法较传统的方法优越。训练后的神经网络模型不仅能准确地拟合训练值,而且能较精确地预测未来趋势。     

Bi-criteria single machine scheduling with a time-dependent learning effect and release times

This paper deals with a bi-criteria single machine scheduling problem with a time-dependent learning effect and release times. The objective is to minimize the weighted sum of the makespan and the total completion time. The problem is NP-hard, thus a mixed integer non-linear programming formulation is presented, and a set of dominance properties are developed. To solve the problem efficiently, a procedure is then proposed by incorporating the dominance properties with an ant colony optimization algorithm. In the proposed algorithm, artificial ants construct solutions as orders of jobs based on the heuristic information as well as pheromone trails. Then, the dominance properties are added to obtain better solutions. To evaluate the algorithm performance, computational experiments are conducted.     

Theorem of learning for a competition algorithm

This paper is an extension of [1], where a new decisive algorithm was proposed. In its operation, the unit resembles artificial neural networks. However, the functioning of the algorithm proposed is based on different concepts. It does not use the concept of a net or a neuron. The theorem of learning for the new competition algorithm is proved.     

Optimization of the Local Search in the Training for SAMANN Neural Network

In this paper, we discuss the visualization of multidimensional data. A well-known procedure for mapping data from a high-dimensional space onto a lower-dimensional one is Sammon’s mapping. This algorithm preserves as well as possible all interpattern distances. We investigate an unsupervised backpropagation algorithm to train a multilayer feed-forward neural network (SAMANN) to perform the Sammon’s nonlinear projection. Sammon mapping has a disadvantage. It lacks generalization, which means that new points cannot be added to the obtained map without recalculating it. The SAMANN network offers the generalization ability of projecting new data, which is not present in the original Sammon’s projection algorithm. To save computation time without losing the mapping quality, we need to select optimal values of control parameters. In our research the emphasis is put on the optimization of the learning rate. The experiments are carried out both on artificial and real data. Two cases have been analyzed: (1) training of the SAMANN network with full data set, (2) retraining of the network when the new data points appear.     

Ant colony optimization with a specialized pheromone trail for the car-sequencing problem

This paper studies the learning process in an ant colony optimization algorithm designed to solve the problem of ordering cars on an assembly line (car-sequencing problem). This problem has been shown to be NP-hard and evokes a great deal of interest among practitioners. Learning in an ant algorithm is achieved by using an artificial pheromone trail, which is a central element of this metaheuristic. Many versions of the algorithm are found in literature, the main distinction among them being the management of the pheromone trail. Nevertheless, few of them seek to perfect learning by modifying the internal structure of the trail. In this paper, a new pheromone trail structure is proposed that is specifically adapted to the type of constraints in the car-sequencing problem. The quality of the results obtained when solving three sets of benchmark problems is superior to that of the best solutions found in literature and shows the efficiency of the specialized trail.     

基于改进的粒子群算法求解供应链网络均衡问题

传统的供应链求解方法为投影法,针对其要对投影进行计算,十分复杂的缺点,提出用改进的粒子群算法求解供应链均衡问题,利用动态异步调整学习因子来有效的提高了算法搜索能力与精度。本文介绍了供应链网络均衡问题转变为无约束优化问题的方法,然后用改进的粒子群优化算法进行求解。通过四个数值算例,将实验结果与标准粒子群算法、蜂群算法、学习因子同步变化的粒子群算法进行比较,验证了改进的粒子群优化算法在解决供应链网络均衡问题中的有效性与优越性,为供应链网络求解提供了一种新的方法。     

A new model and a hyper-heuristic approach for two-dimensional shelf space allocation

In this paper, we propose a two-dimensional shelf space allocation model. The second dimension stems from the height of the shelf. This results in an integer nonlinear programming model with a complex form of objective function. We propose a multiple neighborhood approach which is a hybridization of a simulated annealing algorithm with a hyper-heuristic learning mechanism. Experiments based on empirical data from both real-world and artificial instances show that the shelf space utilization and the resulting sales can be greatly improved when compared with a gradient method. Sensitivity analysis on the input parameters and the shelf space show the benefits of the proposed algorithm both in sales and in robustness.     

Modified optimization layer by layer algorithm for learning multilayer perceptrons

1. IntroductionThe feedforward Multilayer Perceptron (MLP) is one of the most widely used artificial neural networks among other network models. Its field of application includes patternrecognition, identification and control of dynamic systems, system modeling and nonlinearprediction of time series, etc. [1--41 founded on its nonlinear function approximation capability. Research of this type of networks has been stimulated since the discovery andpopularization of the Backpropagation learnin…     

Matrix representation of a binary relation using fuzzy and artificial learning theory. An algorithm which uses the potential functions learning rule

One of the most significant problems in economic domain is the dispose of human preference and choice forecasting. Recently, the economists have focused their researches to use the fuzzy concepts and the artificial learning procedures in the theory of economic choice. This paper extends the work done in this direction and offers a new algorithm for finding the matrix representation of the fuzzy binary relation which describes a preference relation.     

Pattern recognition at different scales: A statistical perspective

In this paper we borrow concepts from Information Theory and Statistical Mechanics to perform a pattern recognition procedure on a set of X-ray hazelnut images. We identify two relevant statistical scales, whose ratio affects the performance of a machine learning algorithm based on statistical observables, and discuss the dependence of such scales on the image resolution. Finally, by averaging the performance of a Support Vector Machines algorithm over a set of training samples, we numerically verify the predicted onset of an “optimal” scale of resolution, at which the pattern recognition is favoured.     

An evolutionary artificial immune system for multi-objective optimization

In this paper, an evolutionary artificial immune system for multi-objective optimization which combines the global search ability of evolutionary algorithms and immune learning of artificial immune systems is proposed. A new selection strategy is developed based upon the concept of clonal selection principle to maintain the balance between exploration and exploitation. In order to maintain a diverse repertoire of antibodies, an information-theoretic based density preservation mechanism is also presented. In addition, the performances of various multi-objective evolutionary algorithms as well as the effectiveness of the proposed features are examined based upon seven benchmark problems characterized by different difficulties in local optimality, non-uniformity, discontinuity, non-convexity, high-dimensionality and constraints. The comparative study shows the effectiveness of the proposed algorithm, which produces solution sets that are highly competitive in terms of convergence, diversity and distribution. Investigations also demonstrate the contribution and robustness of the proposed features.     

Clustering noise-included data by controlling decision errors

Cluster analysis is an unsupervised learning technique for partitioning objects into several clusters. Assuming that noisy objects are included, we propose a soft clustering method which assigns objects that are significantly different from noise into one of the specified number of clusters by controlling decision errors through multiple testing. The parameters of the Gaussian mixture model are estimated from the EM algorithm. Using the estimated probability density function, we formulated a multiple hypothesis testing for the clustering problem, and the positive false discovery rate (pFDR) is calculated as our decision error. The proposed procedure classifies objects into significant data or noise simultaneously according to the specified target pFDR level. When applied to real and artificial data sets, it was able to control the target pFDR reasonably well, offering a satisfactory clustering performance.     

A BB&R algorithm for minimizing total tardiness on a single machine with sequence dependent setup times

This paper presents a Branch, Bound, and Remember (BB&R) exact algorithm using the Cyclic Best First Search (CBFS) exploration strategy for solving the ${1|ST_{sd}|\sum T_{i}}$ scheduling problem, a single machine scheduling problem with sequence dependent setup times where the objective is to find a schedule with minimum total tardiness. The BB&R algorithm incorporates memory-based dominance rules to reduce the solution search space. The algorithm creates schedules in the reverse direction for problems where fewer than half the jobs are expected to be tardy. In addition, a branch and bound algorithm is used to efficiently compute tighter lower bounds for the problem. This paper also presents a counterexample for a previously reported exact algorithm in Luo and Chu (Appl Math Comput 183(1):575–588, 2006) and Luo et?al. (Int J Prod Res 44(17):3367–3378, 2006). Computational experiments demonstrate that the algorithm is two orders of magnitude faster than the fastest exact algorithm that has appeared in the literature. Computational experiments on two sets of benchmark problems demonstrate that the CBFS search exploration strategy can be used as an effective heuristic on problems that are too large to solve to optimality.     

A NEW PERTURBATION SIMPLEX ALGORITHM FOR LINEAR PROGRAMMING

1.IntroductionThedualsimplexalgorithm[1,91andtheprimal-dualsimplealgorithm[6]arewellknownandefficientsimplexvariants.However,bothofthemneedaninitialdualfeasiblebasistogetstarted,andthereforecannotbedirectlyappliedtosolvingproblemsthatdonothavesuchane...     

The optimal combination: Grammatical swarm,particle swarm optimization and neural networks

Social behaviour is mainly based on swarm colonies, in which each individual shares its knowledge about the environment with other individuals to get optimal solutions. Such co-operative model differs from competitive models in the way that individuals die and are born by combining information of alive ones. This paper presents the particle swarm optimization with differential evolution algorithm in order to train a neural network instead the classic back propagation algorithm. The performance of a neural network for particular problems is critically dependant on the choice of the processing elements, the net architecture and the learning algorithm. This work is focused in the development of methods for the evolutionary design of artificial neural networks. This paper focuses in optimizing the topology and structure of connectivity for these networks.     

Hybrid chaos optimization algorithm with artificial emotion

Much research on Artificial Intelligence (AI) has been focusing on exploring various potential applications of intelligent systems. In most cases, the researches attempt to model human intelligence by mimicking the brain structure and function, but they ignore an important aspect in human learning and decision making: the artificial emotion. In this paper, we present a new unconstrained global optimization method, hybrid chaos optimization algorithm with artificial emotion (HCOAAE), which avoids trapping to local minima, and improves convergence in large space and high-dimension optimization problems. The main purpose of artificial emotion is to mimic decision making behavior process of humans, to choose most suitable parameters of HCOAAE and decide whether to change current search strategy or not in the next iteration. Numerical simulations of 13 benchmark functions with different dimensions are used to test the performance of HCOAAE. Experimental results show that the proposed method significantly outperforms the existing methods in terms of convergence speed, computational effectiveness, and numerical stability.     

基于自适应均匀变异人工鱼群投资组合优化算法的研究

投资市场具有一定的风险,影响因素包括经济、政治、市场自身规律等,根据市场机制构建合适的投资组合模型,可以有效降低市场风险,提高投资回报率.人工鱼群算法是模仿自然界鱼类的一种人工智能优化算法,具有较好的优化能力,但有时会陷入局部最优解.首先将人工鱼群算法与均匀变异相结合,加入均匀变异随机数,使算法能够跳出局部最优解,得到全局最优,从而提高算法精度.然后采用改进人工鱼群算法对投资组合模型进行优化求解.实验表明,改进人工鱼群算法具有较好的收敛精度和收敛速度,对投资组合模型的求解效果更好,风险下降,收益增加、     

高维约束矩阵回归问题

高维约束矩阵回归是指高维情况下带非凸约束的多响应多预测统计回归问题,其数学模型是一个NP-难的矩阵优化,它在机器学习与人工智能、医学影像疾病诊疗、基因表达分析、脑神经网络、风险管理等领域有广泛应用.从高维约束矩阵回归的优化理论和算法两方面总结和评述这些新成果,同时,列出了相应的重要文献.     

检测网络社团结构的基于最短路径的$k$-均值算法

我们考虑复杂网络社团结构的检测问题,即检测出那些具有高于平均密度的边所连接的节点的集合.本文我们利用模拟退火策略来极大化可表示为稳定效益函数的模量(modularity),并结合基于最短路径的$k$-均值迭代过程来对网络进行分区.该算法不仅能检测出社团,而且能够识别出在最短路径度量下,该社团中位于中心位置的节点.社团的最优数目可以在无需任何关于网络结构的先验信息下自动确定.对人工生成网络和真实世界中的网络的成功应用表明了算法的有效性.     

A sequential predictor retraining algorithm and its application to market prediction

In many applications, there are a large number of predictors, designed manually or trained automatically, to predict the same outcome. Much research has been devoted to the design of algorithms that can effectively select/combine these predictors to generate a more accurate ensemble predictor. The collaborative training algorithms from attribute distributed learning provide batch-processing solutions for scenarios in which the individual predictors are heterogeneous, taking different inputs and employing different models. However, in some applications, for example financial market prediction, it is desirable to use an online approach. In this paper, an innovative online algorithm is proposed, stemming from the collaborative training algorithms developed for attribute distributed learning. It sequentially takes new observations, simultaneously adjusting the way that the individual predictors are combined, and provides feedback to the individual predictors for them to be retrained in order to achieve a better ensemble predictor in real time. The efficacy of this new algorithm is demonstrated by extensive simulations on both artificial and real data, and particularly for financial market data. A trading strategy constructed from the ensemble predictor shows strong performance when applied to financial market prediction.