Sirapat Chiewchanwattana

Robust extreme learning machine

The output weights computing of extreme learning machine (ELM) encounters two problems, the computational and outlier robustness problems. The computational problem occurs when the hidden layer output matrix is a not full column rank matrix or an ill-conditioned matrix because of randomly generated input weights and biases. An existing solution to this problem is Singular Value Decomposition (SVD) method. However, the training speed is still affected by the large complexity of SVD when computing the Moore-Penrose (MP) pseudo inverse. The outlier robustness problem may occur when the training data set contaminated with outliers then the accuracy rate of ELM is extremely affected. This paper proposes the Extended Complete Orthogonal Decomposition (ECOD) method to solve the computational problem in ELM weights computing via ECODLS algorithm. And the paper also proposes the other three algorithms, i.e. the iteratively reweighted least squares (IRWLS-ELM), ELM based on the multivariate least-trimmed squares (MLTS-ELM), and ELM based on the one-step reweighted MLTS (RMLTS-ELM) to solve the outlier robustness problem. However, they also encounter the computational problem. Therefore, the ECOD via ECODLS algorithm is also used successfully in the three proposed algorithms. The experiments of regression problems were conducted on both toy and real-world data sets. The outlier types are one-sided and two-sided outliers. Each experiment was randomly contaminated with outliers, of one type only, with 10%, 20%, 30%, 40%, and 50% of the total training data size. Meta-metrics evaluation was used to measure the outlier robustness of the proposed algorithms compared to the existing algorithms, i.e. the minimax probability machine regression (MPMR) and the ordinary ELM. The experimental results showed that ECOD can effectively replace SVD. The ECOD is robust to the not full column rank or the ill-conditional problem. The speed of the ELM training using ECOD is also faster than the ordinary training algorithm. Moreover, the meta-metrics measure showed that the proposed algorithms are less affected by the increasing number of outliers than the existing algorithms.

An Improved Grey Wolf Optimizer for Training q-Gaussian Radial Basis Functional-link Nets

In this paper, a novel meta-heuristic technique an improved Grey Wolf Optimizer (IGWO) which is an improved version of Grey Wolf Optimizer (GWO) is proposed. The performance is evaluated by adopting the IGWO to training q-Gaussian Radial Basis Functional-link nets (qRBFLNs) neural networks. The function approximation problems in regression areas and the multiclass classification problem in classification areas are employed to test the algorithm. For instance, in order to overcome the multiclass classification problem, the dataset of the screening risk groups of the population age 15 years and over in Charoensin District, Sakon Nakhon Province, Thailand is used in the experiments. The results of the function approximation problems and real application in multiclass classification problem prove that the proposed algorithm is able to address the test problems. Moreover, the proposed algorithm obtains competitive performance compared to other meta-heuristic methods.

A Comparative Study of Improved Artificial Bee Colony Algorithms Applied to Multilevel Image Thresholding

Multilevel thresholding is a highly useful tool for the application of image segmentation. Otsu’s method, a common exhaustive search for finding optimal thresholds, involves a high computational cost. There has been a lot of recent research into various meta-heuristic searches in the area of optimization research. This paper analyses and discusses using a family of artificial bee colony algorithms, namely, the standard ABC, ABC/best/1, ABC/best/2, IABC/best/1, IABC/rand/1, and CABC, and some particle swarm optimization-based algorithms for searching multilevel thresholding. The strategy for an onlooker bee to select an employee bee was modified to serve our purposes. The metric measures, which are used to compare the algorithms, are the maximum number of function calls, successful rate, and successful performance. The ranking was performed by Friedman ranks. The experimental results showed that IABC/best/1 outperformed the other techniques when all of them were applied to multilevel image thresholding. Furthermore, the experiments confirmed that IABC/best/1 is a simple, general, and high performance algorithm.

Imputing incomplete time-series data based on varied-window similarity measure of data sequences

This paper presents a pattern characterization approach for the imputation of missing samples of time-series data. The new algorithm is based on the observation that time-series data that are manifestations of natural phenomena contain several sets of similar time-series subsequences. The imputation of missing samples is achieved by finding a complete subsequence that is similar to the missing sample subsequence and imputing the missing samples from this complete subsequence. The new algorithm is tested using standard benchmark as well as real-world data sets. The experimental results showed that the imputation accuracy of the proposed algorithm, referred to as the varied-window similarity measure (VWSM) algorithm, is comparable or better than traditional methods such as: the spline interpolation, the multiple imputation (MI), and the optimal completion strategy fuzzy c-means algorithm (OCSFCM) in case of non-stationary time-series data.

Time-series data prediction based on reconstruction of missing samples and selective ensembling of FIR neural networks

This paper considers the problem of time-series forecasting by a selective ensemble neural network when the input data are incomplete. Five fill-in methods, viz. cubic smoothing spline interpolation, EM (Expectation maximization), regularized EM, average EM, and average regularized EM, are simultaneously employed in a first step for reconstructing the missing values of time-series data. A set of complete data from each individual fill-in method is used to train a finite impulse response (FIR) neural network to predict the time series. The outputs from individual network are combined by a selective ensemble method in the second step. Experimental results show that the prediction made by the proposed method is more accurate than those predicted by neural networks without a fill-in process or by a single fill-in process.

Improved differential evolution via cuckoo search operator

Differential Evolution (DE) is a very popular optimization algorithm for solving numerical optimization problems. It is simple yet powerful algorithm, which has shown effective performance in many optimization problems. In this paper, DECSO that uses the Abandon operator of Cuckoo search to improve the exploration ability of the original DE was proposed. The experimental studies on ten well-known benchmark functions have shown that the proposed approach has efficient search power and fast convergence.

A Global Multilevel Thresholding Using Differential Evolution Approach

Otsu’s function measures the properness of threshold values in multilevel image thresholding. Optimal threshold values are necessary for some applications and a global search algorithm is required. Differential evolution (DE) is an algorithm that has been used successfully for solving this problem. Because the difficulty of a problem grows exponentially when the number of thresholds increases, the ordinary DE fails when the number of thresholds is greater than 12. An improved DE, using a new mutation strategy, is proposed to overcome this problem. Experiments were conducted on 20 real images and the number of thresholds varied from 2 to 16. Existing global optimization algorithms were compared with the proposed algorithms, that is, DE, rank-DE, artificial bee colony (ABC), particle swarm optimization (PSO), DPSO, and FODPSO. The experimental results show that the proposed algorithm not only achieves a more successful rate but also yields a lower threshold value distortion than its competitors in the search for optimal threshold values, especially when the number of thresholds is large.

Enhancing modified cuckoo search by using Mantegna Lévy flights and chaotic sequences

In this paper, we present an improvement of the modified cuckoo search (MCS) method. We focus on a new nest generation from the top nest group. This group of nests assists a better local search. We use Tent map chaotic sequences to replace the constant parameter, inverse golden ratio of MCS. This process aims to find a better solution in case of multi-modal problems. The Cauchy Lévy distribution is replaced by Mantegna Lévy distribution generation. This process assists in finding a better solution in case of uni-modal problems. To construct a more suitable with wider optimization problems and good convergence property, these two concepts are combined together as Improved MCS with Chaotic Sequences algorithm (ICMCS). The proposed algorithm is verified using nineteen constrained optimization problems. The performance of the proposed algorithm is compared with the original MCS algorithms. The optimal solutions obtained in this study are more superior to MCS.

Evolutionary Circular Extreme Learning Machine

Circular Extreme Learning Machine (C-ELM) is an extension of Extreme Learning Machine. Its power is mapping both linear and circular separation boundaries. However, C-ELM uses the random determination of the input weights and hidden biases, which may lead to local optimal. This paper proposes a hybrid learning algorithms based on the C-ELM and the Differential Evolution (DE) to select appropriate weights and hidden biases. It called Evolutionary circular extreme learning machine (EC-ELM). From experimental results show EC-ELM can slightly improve C-ELM and also reduce the number of nodes network.

R r-cr -IJADE: An efficient differential evolution algorithm for multilevel image thresholding

Abstract There is a need for a new method of segmentation to improve the efficiency of expert systems that need segmentation. Multilevel thresholding is a widely used technique that uses threshold values for image segmentation. However, from a computational stand point, the search for optimal threshold values presents a challenging task, especially when the number of thresholds is high. To get the optimal threshold values, a meta-heuristic or optimization algorithm is required. Our proposed algorithm is referred to as Rr-cr-IJADE, which is an improved version of Rcr-IJADE. Rr-cr-IJADE uses a newly proposed mutation strategy, “DE/rand-to-rank/1”, to improve the search success rate. The strategy uses the parameter F adaptation, crossover rate repairing, and the direction from a randomly selected individual to a ranking-based leader. The complexity of the proposed algorithm does not increase, compared to its ancestor. The performance of Rr-cr-IJADE, using Otsus function as the objective function, was evaluated and compared with other state-of-the-art evolutionary algorithms (EAs) and swarm intelligence algorithms (SIs), under both ‘low-level’ and ‘high-level’ experimental sets. Within the ‘low-level’ sets, the number of thresholds varied from 2 to 16, within 20 real images. For the ‘high-level’ sets, the threshold numbers chosen were 24, 32, 40, 48, 56 and 64, within 2 synthetic pseudo images, 7 satellite images, and three real images taken from the set of 20 real images. The proposed Rr-cr-IJADE achieved higher success rates with lower threshold value distortion (TVD) than the other state-of-the-art EA and SI algorithms.