Fuzzy decision fusion of complementary experts based on evolutionary cepstral coefficients for phoneme recognition

Optimal representation of acoustic features is an ongoing challenge in automatic speech recognition research. As an initial step toward this purpose, optimization of filterbanks for the cepstral coefficient using evolutionary optimization methods is proposed in some approaches. However, the large number of optimization parameters required by a filterbank makes it difficult to guarantee that an individual optimized filterbank can provide the best representation for phoneme classification. Moreover, in many cases, a number of potential solutions are obtained. Each solution presents discrimination between specific groups of phonemes. In other words, each filterbank has its own particular advantage. Therefore, the aggregation of the discriminative information provided by filterbanks is demanding challenging task. In this study, the optimization of a number of complementary filterbanks is considered to provide a different representation of speech signals for phoneme classification using the hidden Markov model (HMM). Fuzzy information fusion is used to aggregate the decisions provided by HMMs. Fuzzy theory can effectively handle the uncertainties of classifiers trained with different representations of speech data. In this study, the output of the HMM classifiers of each expert is fused using a fuzzy decision fusion scheme. The decision fusion employed a global and local confidence measurement to formulate the reliability of each classifier based on both the global and local context when making overall decisions. Experiments were conducted based on clean and noisy phonetic samples. The proposed method outperformed conventional Mel frequency cepstral coefficients under both conditions in terms of overall phoneme classification accuracy. The fuzzy fusion scheme was shown to be capable of the aggregation of complementary information provided by each filterbank.     

A novel artificial bee colony algorithm with Powell's method

Artificial bee colony (ABC) algorithm is a relatively new optimization technique which has been shown to be competitive to other population-based algorithms. However, there is still an insufficiency in ABC regarding its solution search equation, which is good at exploration but poor at exploitation. To address this concerning issue, we first propose a modified search equation which is applied to generate a candidate solution in the onlookers phase to improve the search ability of ABC. Further, we use the Powell's method as a local search tool to enhance the exploitation of the algorithm. The new algorithm is tested on 22 unconstrained benchmark functions and 13 constrained benchmark functions, and are compared with some other ABCs and several state-of-the-art algorithms. The comparisons show that the proposed algorithm offers the highest solution quality, fastest global convergence, and strongest robustness among all the contenders on almost all test functions.     

Parameter estimation for five- and seven-parameter photovoltaic electrical models using evolutionary algorithms

Equivalent electric circuit modeling of PV devices is widely used to predict PV electrical performance. The first task in using the model to calculate the electrical characteristics of a PV device is to find the model parameters which represent the PV device. In the present work, parameter estimation for the model parameter using various evolutionary algorithms is presented and compared. The constraint set on the estimation process is that only the data directly available in module datasheets can be used for estimating the parameters. The electrical model accuracy using the estimated parameters is then compared to several electrical models reported in literature for various PV cell technologies.     

From Constrained to Unconstrained Maximum Agreement Subtree in Linear Time

We propose and study the Maximum Constrained Agreement Subtree (MCAST) problem, which is a variant of the classical Maximum Agreement Subtree (MAST) problem. Our problem allows users to apply their domain knowledge to control the construction of the agreement subtrees in order to get better results. We show that the MCAST problem can be reduced to the MAST problem in linear time and thus we have algorithms for MCAST with running times matching the fastest known algorithms for MAST. A preliminary version of this paper appears in the Proceedings of the Fifth Workshop on Algorithms in Bioinformatics (WABI 2005). Research of H.F. Ting is supported in part by Hong Kong RGC Grant HKU-7172/06E.     

Evolving BlenX programs to simulate the evolution of biological networks

We present a formal approach to study the evolution of biological networks. We use the Beta Workbench and its BlenX language to model and simulate networks in connection with evolutionary algorithms. Mutations are done on the structure of BlenX programs and networks are selected at any generation by using a fitness function. The feasibility of the approach is illustrated with a simple example.     

Expected runtimes of evolutionary algorithms for the Eulerian cycle problem

Evolutionary algorithms are randomized search heuristics, which are applied to problems whose structure is not well understood, as well as to problems in combinatorial optimization. They have successfully been applied to different kinds of arc routing problems. To start the analysis of evolutionary algorithms with respect to the expected optimization time on these problems, we consider the Eulerian cycle problem. We show that a variant of the well-known (1+1)$(1+1)$ EA working on the important encoding of permutations is able to find an Eulerian tour of an Eulerian graph in expected polynomial time. Altering the operator used for mutation in the considered algorithm, the expected optimization time changes from polynomial to exponential.     

An optimized modular neural network controller based on environment classification and selective sensor usage for mobile robot reactive navigation

A new approach to the design of a neural network (NN) based navigator is proposed in which the mobile robot travels to a pre-defined goal position safely and efficiently without any prior map of the environment. This navigator can be optimized for any user-defined objective function through the use of an evolutionary algorithm. The motivation of this research is to develop an efficient methodology for general goal-directed navigation in generic indoor environments as opposed to learning specialized primitive behaviors in a limited environment. To this end, a modular NN has been employed to achieve the necessary generalization capability across a variety of indoor environments. Herein, each NN module takes charge of navigating in a specialized local environment, which is the result of decomposing the whole path into a sequence of local paths through clustering of all the possible environments. We verify the efficacy of the proposed algorithm over a variety of both simulated and real unstructured indoor environments using our autonomous mobile robot platform.     

改进的粒子群优化算法

将基本粒子群算法粒子行为基于个体极值点和全局极值点变化为基于个体极值中心,并且按一定概率选择其他粒子的个体极值点,设计了一种新的粒子群优化算法.新算法的学习行为符合自然界生物的学习规律,更有利于粒子发现问题的全局最优解.实验结果表明了算法的有效性.     

一种特征加权FCM算法的图像重建技术研究

针对传统FCM(Fuzzy C-means)算法中初始聚类中心选取的随机性以及对初始值敏感的问题,提出一种基于进化策略的色彩空间加权的FCM聚类算法.通过在RGB(Red Green Blue)色彩空间矢量中设置加权矩阵来补偿各色彩的非均匀性,并采用一种类内最小距离最大的统计聚类算法来初始化聚类中心.实验结果表明,该算法能有效减少颜色量化后的均方差值,保持重建图像的整体层次和局部特征细节,对研究图像处理技术有较强的实际意义.

Abstract：

Aiming at a defect on randomness of the initial clustering center choosing and sensitivity of initial value in tradition FCM(fuzzy C-means) algorithm, a clustering algorithm about FCM of weighted color space based on evolutionary strategy is proposed. By interposing weighted matrix in RGB(Red Green Blue) color space, the color's inhomogeneous is compensated. And by using a statistics clustering algorithm of minimal maximal distance, clustering center is initiated. The experimental results show that the algorithm can decrease effectively the mean square deviation of color quantization, keep overall arrangement of ideas and part characteristic detail in image reconstruction, and has practical value to the study of the image process technology.     

一种实数编码量子进化算法及其收敛性

基于量子计算理论和进化理论,提出一种新的量子进化算法--基于实数编码的量子进化算法(RQEA).不同于传统进化算法的单点编码和量子进化算法的量子比特编码,该算法以实数矩形区域表示基因,一条染色体携带多个个体信息,利用量子态叠加和相干机理,通过叠加、变异及自学习来完成进化过程,理论分析证明了算法具有全局收敛性,实验结果表明,该算法在函数优化上具有优异的性能.