Optimal weight tuning method for unit selection cost functions in syllable based text-to-speech synthesis

This paper proposes a method for tuning the weights of unit selection cost functions in syllable based text-to-speech (TTS) synthesis system. In this work, unit selection cost functions, namely target cost and concatenation cost, are designed appropriate to syllables. The method tunes the weights in such a way that perceptual preference patterns are appropriately considered while selecting the units. The method uses genetic algorithm to derive the optimal weights. Fitness function is designed to map perceptual preference patterns into weights of unit selection cost functions. The effectiveness of proposed method is evaluated by both subjective and objective measures. From the results, it is observed that the derived optimal weights can synthesize good quality speech compared to manually tuned weights.     

一种求解函数优化问题的改进鲸鱼优化算法

为提高鲸鱼优化算法求解复杂函数优化问题的性能,提出一种基于自适应参数及小生境技术的改进鲸鱼优化算法。首先,引入自适应概率阈值协调算法的全局探索及局部开发能力;其次,利用自适应位置权重对鲸鱼位置更新公式进行调整,提高算法的收敛速度及寻优精度;最后,采用预选择小生境技术,避免算法出现早熟收敛的现象。通过对12个典型基准测试函数的仿真表明,改进算法的寻优精度和收敛速度较对比算法均有明显提升,证明了提出的改进策略能有效提高鲸鱼优化算法求解复杂函数优化问题的性能。     

Measuring speech quality for text-to-speech systems: development and assessment of a modified mean opinion score (MOS) scale

The quality of text-to-speech systems can be effectively assessed only on the basis of reliable and valid listening tests to assess overall system performance. A mean opinion scale (MOS) has been the recommended measure of synthesized speech quality [ITU-T Recommendation P.85, 1994. Telephone transmission quality subjective opinion tests. A method for subjective performance assessment of the quality of speech voice output devices]. We assessed this MOS scale and developed and tested a modified measure of speech quality. This modified measure has new items specific to text-to-speech systems. Our research was motivated by the lack of clear evidence of the conceptual content of as well as the psychometric properties of the MOS scale. We present conceptual arguments and empirical evidence for the reliability and validity of a modified scale. Moreover, we employ state of the art psychometric techniques such as confirmatory factor analysis to provide strong tests of psychometric properties. This modified scale is better suited to appraise synthesis systems since it includes items that are specific to the artifacts found in synthesized speech. We believe that the speech synthesis research communities will find this modified scale a better fit for listening tests to assess synthesized speech.     

Multi-objective community detection in complex networks

Community detection in social network analysis is usually considered as a single objective optimization problem, in which different heuristics or approximate algorithms are employed to optimize a objective function that capture the notion of community. Due to the inadequacy of those single-objective solutions, this paper first formulates a multi-objective framework for community detection and proposes a multi-objective evolutionary algorithm for finding efficient solutions under the framework. After analyzing and comparing a variety of objective functions that have been used or can potentially be used for community detection, this paper exploits the concept of correlation between objective which charcterizes the relationship between any two objective functions. Through extensive experiments on both artifical and real networks, this paper demonstrates that a combination of two negatively correlated objectives under the multi-objective framework usually leads to remarkably better performance compared with either of the orignal single objectives, including even many popular algorithms..     

多目标最优化云工作流调度进化遗传算法

为了实现云环境中科学工作流调度的执行跨度和执行代价的同步优化,提出了一种多目标最优化进化遗传调度算法MOEGA。该算法以进化遗传为基础,定义了任务与虚拟机映射、虚拟机与主机部署间的编码机制,设计了满足多目标优化的适应度函数。同时,为了满足种群的多样性,在调度方案中引入了交叉与变异操作,并使用启发式方法进行种群初始化。通过4种现实科学工作流的仿真实验,将其与同类型算法进行了性能比较。结果表明,MOEGA算法不仅可以满足工作流截止时间约束,而且在降低任务执行跨度与执行代价的综合性能方面也优于其他算法。     

Accelerating Differential Evolution Using an Adaptive Local Search

We propose a crossover-based adaptive local search (LS) operation for enhancing the performance of standard differential evolution (DE) algorithm. Incorporating LS heuristics is often very useful in designing an effective evolutionary algorithm for global optimization. However, determining a single LS length that can serve for a wide range of problems is a critical issue. We present a LS technique to solve this problem by adaptively adjusting the length of the search, using a hill-climbing heuristic. The emphasis of this paper is to demonstrate how this LS scheme can improve the performance of DE. Experimenting with a wide range of benchmark functions, we show that the proposed new version of DE, with the adaptive LS, performs better, or at least comparably, to classic DE algorithm. Performance comparisons with other LS heuristics and with some other well-known evolutionary algorithms from literature are also presented.     

Free lunches on the discrete Lipschitz class

The No-Free-Lunch theorem states that there does not exist a genuine general-purpose optimizer because all algorithms have the identical performance on average over all functions. However, such a result does not imply that search heuristics or optimization algorithms are futile if we are more cautious with the applicability of these methods and the search space. In this paper, within the No-Free-Lunch framework, we firstly introduce the discrete Lipschitz class by transferring the Lipschitz functions, i.e., functions with bounded slope, as a measure to fulfill the notion of continuity in discrete functions. We then investigate the properties of the discrete Lipschitz class, generalize an algorithm called subthreshold-seeker for optimization, and show that the generalized subthreshold-seeker outperforms random search on this class. Finally, we propose a tractable sampling-test scheme to empirically demonstrate the superiority of the generalized subthreshold-seeker under practical configurations. This study concludes that there exist algorithms outperforming random search on the discrete Lipschitz class in both theoretical and practical aspects and indicates that the effectiveness of search heuristics may not be universal but still general in some broad sense.     

基于鱼群优化算法和Cholesky分解的RELM的基因表达数据分类

提出一种基于鱼群优化算法和Cholesky分解的改进的正则极限学习机算法(FSC-RELM)来对基因表达数据进行分类。FSC-RELM算法中,首先用鱼群优化算法对RELM输入层权值进行优化,其中目标函数定义为误差函数的倒数;再对RELM输出层权值矩阵进行分解,采用Cholesky分解法进行优化,以提高算法速度,减少训练时间。为了评价算法性能,对若干标准基因数据集进行了实验,结果表明,FSC-RELM算法在较短的时间内可以获得较高的分类精度,性能优异。     

基于串联协作的多方法协作优化方法

为更好地求解复杂优化问题的全局最优解,提出了基于串联协作的多方法协作优化方法。它采用串联方式组织各个优化方法之间的协作。各个优化方法在每次协作中进行若干次优化,每一优化方法在优化过程中及优化结果中包含的优化信息被用于其它优化方法中,以提高整体优化性能。选择了遗传算法、模式搜索法和Powell法等三个直接优化方法进行串联协作组成的多方法协作优化方法的设计,并对其优化特性进行了深入的理论分析。最后,用三个复杂多维函数对算法性能进行了测试。计算实例表明,基于串联协作的多方法协作优化方法取得了优于单个优化方法的全局最优特性。     

结合文化算法的多种群协同变异PSO算法

粒子群算法是一种新的基于群体智能的启发式全局优化算法,其概念简单,易于实现,而且具有良好的优化性能,目前已在许多领域得到应用。但在求解高维多峰函数寻优问题时,算法易陷入局部最优。结合文化算法和高斯变异的思想,提出一种基于文化算法和高斯变异的多群协同粒子群算法。该算法可以摆脱局部最优解对微粒的吸引,基于典型高维复杂函数的仿真结果表明,与多种群粒子群优化算法相比,该混合算法具有更好的优化性能。     

High speed ant colony optimization CMOS chip

Ant colony optimization (ACO) is an optimization computation inspired by the study of the ant colonies’ behavior. This paper presents design and CMOS implementation of the ant colony optimization based algorithm for solving the TSP problem. In order to implement ant colony optimization algorithm in CMOS, we will present a new algorithm. This algorithm is based on the original ant colony optimization but it can be implemented in CMOS. Briefly, pheromone matrix is transformed on the chip area and ants move up-down through the pheromone matrix and they make their decisions. Finally ants select a global path. In previous researches only pheromone values is used, but select the next city in this paper is based on heuristics value and pheromone value. In definition of problem, we use heuristics value as a matrix. Previous researches could not be used for wide type of optimization problem but our chip gives heuristics value initially and we can change initial value of heuristics value according to the optimization problem so this capability increases the flexibility of ACO chip. Simple circuit is used in blocks of our chip to increase the speed of convergence of ACO chip. We use Linear Feedback Shift Register (LSFR) circuit for random number generator in ACO chip. ACO chip has capability of solving the big TSP problem. ACO chip is simulated by HSPICE software and simulation results show the good performance of final chip.     

Cost-sensitive label embedding for multi-label classification

Label embedding (LE) is an important family of multi-label classification algorithms that digest the label information jointly for better performance. Different real-world applications evaluate performance by different cost functions of interest. Current LE algorithms often aim to optimize one specific cost function, but they can suffer from bad performance with respect to other cost functions. In this paper, we resolve the performance issue by proposing a novel cost-sensitive LE algorithm that takes the cost function of interest into account. The proposed algorithm, cost-sensitive label embedding with multidimensional scaling (CLEMS), approximates the cost information with the distances of the embedded vectors by using the classic multidimensional scaling approach for manifold learning. CLEMS is able to deal with both symmetric and asymmetric cost functions, and effectively makes cost-sensitive decisions by nearest-neighbor decoding within the embedded vectors. We derive theoretical results that justify how CLEMS achieves the desired cost-sensitivity. Furthermore, extensive experimental results demonstrate that CLEMS is significantly better than a wide spectrum of existing LE algorithms and state-of-the-art cost-sensitive algorithms across different cost functions.     

组播路由综合优化及其算法

路由优化是组播通信的一个关键问题,它的目标是将低路由费用。论文中首次提出了组播路由费用综合优化的思想,即组播路由的网络费用和目的地费用同时得到优化：优化网络费用以降低路由消耗的网络资源;优化目的地费用以减少源到目的地的平均时延。并提出了一种组播路由综合优化算法,该算法的时间复杂性低,实现简单,还可以分布式实现。     

融合振幅随机补偿与步长演变机制的改进原子搜索优化算法

针对原子优化算法寻优精度弱且易陷入局部极值的问题,本文从种群多样性、参数适应性和位置动态性角度提出一种融合混沌优化、振幅随机补偿和步长演变机制改进的原子搜索优化算法(improved atom search optimization, IASO),并将其成功应用于分类任务。首先,引入帐篷映射(Tent混沌)增强原子种群在搜索空间中的分布均匀性;其次,通过构建振幅函数对算法参数进行随机扰动并加入步长演变因子更新原子位置,以增强算法全局性和收敛性;最后,再将改进算法应用于误差反馈神经网络(BP神经网络)参数优化。通过与6种元启发式算法在20个基准测试函数下的数值实验对比表明:IASO不仅在求解多维基准函数上具有好的寻优性能,且在对BP神经网络参数进行优化时相较于2种对比算法具有更高的分类精度。     

Radial basis function networks, regression weights, and theexpectation-maximization algorithm

We propose a modified radial basis function (RBF) network in which the regression weights are used to replace the constant weights in the output layer. It is shown that the modified RBF network can reduce the number of hidden units significantly. A computationally efficient algorithm, known as the expectation-maximization (EM) algorithm, is used to estimate the parameters of the regression weights. A salient feature of this algorithm is that it decomposes a complicated multiparameter optimization problem into L separate small-scale optimization problems, where L is the number of hidden units. The superior performance of the modified RB network over the standard RBF network is illustrated by computer simulations     

类搜索算法

提出利用类结构驱动的群体进化计算方法——类搜索算法(CSA).CSA在个体间构造簇类形态的虚拟连接关系,并通过对类组织的结构和类搜索过程进行动态调节来优化模拟进化系统的计算状态,提高群体的搜索效率.介绍了CSA的基本模型,并基于CSA融合进化算子与差分计算机制设计出数值优化算法CSA/DE.对多个典型高纬函数和复杂混合函数的仿真实验结果说明,CSA/DE是一种对高纬连续问题高效、稳定的搜索优化方法.该工作一方面验证了CSA的可行性和有效性;另一方面则显示:基于类搜索模型可有效融合异构且具有不同计算特性的搜索机制,形成对待求解问题更具针对性且协调性更佳的搜索计算方法.这为高性能优化算法的设计提供了一条新的途径.     

随机微粒群优化算法

微粒群优化算法是继蚁群算法之后又一种新的基于群体智能的启发式全局优化算法,其概念简单、易于实现,而且具有良好的优化性能,目前已在许多领域得到应用。但在求解高维多峰函数寻优问题时,算法易陷入局部最优。该文结合模拟退火算法的思想,提出了一种改进的微粒群优化算法——随机微粒群优化算法,该算法在运行初期具有更强的探索能力,可以避免群体过早陷入局部极值点。基于典型高维复杂函数的仿真结果表明,与基本微粒群优化算法相比,该混合算法具有更好的优化性能。     

On the investigation of hyper-heuristics on a financial forecasting problem

Financial forecasting is a really important area in computational finance, with numerous works in the literature. This importance can be reflected in the literature by the continuous development of new algorithms. Hyper-heuristics have been successfully used in the past for a number of search and optimization problems, and have shown very promising results. To the best of our knowledge, they have not been used for financial forecasting. In this paper we present pioneer work, where we use different hyper-heuristics frameworks to investigate whether we can improve the performance of a financial forecasting tool called EDDIE 8. EDDIE 8 allows the GP (Genetic Programming) to search in the search space of indicators for solutions, instead of using pre-specified ones; as a result, its search area has dramatically increased and sometimes solutions can be missed due to ineffective search. We apply 14 different low-level heuristics to EDDIE 8, to 30 different datasets, and examine their effect to the algorithm’s performance. We then select the most prominent heuristics and combine them into three different hyper-heuristics frameworks. Results show that all three frameworks are competitive, and are able to show significantly improved results, especially in the case of best results. Lastly, analysis on the weights of the heuristics shows that there can be a constant swinging among some of the low-level heuristics, which denotes that the hyper-heuristics frameworks are able to ‘know’ the appropriate time to switch from one heuristic to the other, based on their effectiveness.     

基于凸优化的3D变迹函数水下声呐成像算法

针对水下声呐成像算法中旁瓣电平过高的问题,提出一种基于凸优化的3D变迹函数优化方法,用于成像算法之前,以求在波束形成中获得较低旁瓣电平。利用凸优化算法高效求解满足约束条件的阵元激励权值,将激励权值形成一个3D变迹函数,以幅度变迹的方式作用于各阵元接收的信号,并对幅度变迹后的各阵元接收信号采用延时叠加方法波束形成并成像。以主瓣宽度和峰值旁瓣电平作为衡量波束性能的指标,将所提出方法与切比雪夫窗函数方法和未经变迹函数方法处理进行对比,并进行成像仿真。实验结果表明:所提出方法可以有效地抑制波束峰值旁瓣电平,提高成像对比度。     

Probabilistic opposition-based particle swarm optimization with velocity clamping

A probabilistic opposition-based Particle Swarm Optimization algorithm with Velocity Clamping and inertia weights (OvcPSO) is designed for function optimization—to accelerate the convergence speed and to optimize solution’s accuracy on standard benchmark functions. In this work, probabilistic opposition-based learning for particles is incorporated with PSO to enhance the convergence rate—it uses velocity clamping and inertia weights to control the position, speed and direction of particles to avoid premature convergence. A comprehensive set of 58 complex benchmark functions including a wide range of dimensions have been used for experimental verification. It is evident from the results that OvcPSO can deal with complex optimization problems effectively and efficiently. A series of experiments have been performed to investigate the influence of population size and dimensions upon the performance of different PSO variants. It also outperforms FDR-PSO, CLPSO, FIPS, CPSO-H and GOPSO on various benchmark functions. Last but not the least, OvcPSO has also been compared with opposition-based differential evolution (ODE); it outperforms ODE on lower swarm population and higher-dimensional functions.