Search

Convolutional neural networks (CNNs) are deep learning models used widely for solving various tasks like computer vision and speech recognition. CNNs are developed manually based on problem-specific domain knowledge and tricky settings, which are laborious, time consuming, and challenging. To solve these, our study develops an improved differential evolution of convolutional neural network (IDECNN) algorithm to design CNN layer architectures for image classification. Variable-length encoding is utilized to represent the flexible layer architecture of a CNN model in IDECNN. An efficient heuristic mechanism is proposed in IDECNN to evolve CNN architecture through mutation and crossover to prevent premature convergence during the evolutionary process. Eight well-known imaging datasets were utilized. The results showed that IDECNN could design suitable architecture compared with 20 existing CNN models. Finally, CNN architectures are applied to pneumonia and coronavirus disease 2019 (COVID-19) X-ray biomedical image data. The results demonstrated the usefulness of the proposed approach to generate a suitable CNN model.

Neural Architecture Search (NAS) is an automatic process of designing a neural architecture for solving classification problems. It is closely related to hyper-parameters such as hidden layers, neurons in each hidden layer, type of activation function (ACT), network optimizer and so on. Therefore, finding appropriate hyper-parameters to construct suitable network architecture for a particular problem is a challenging task. In this paper, an improved Genetic Algorithm (GA-NAS) is proposed to build a multi-layer feed forward architecture for image classification problem. Each chromosome of the proposed method is encoded with four hyper-parameters namely no. of hidden layers, neurons per hidden layer, activation function (ACT) and network error optimization technique. Each chromosome represents a neural network architecture for the given problem. The categorical cross-entropy or log function is considered to represent fitness function which provides performance accuracy of the architecture. The proposed methodology is experimented on two well-known benchmark image classification data sets such as CIFAR-10 and MNIST. The GA-NAS is compared with brute force algorithm and obtained results demonstrated the effectiveness for solving image classification problems.

Activation functions in the neural networks play an important role by introducing non-linear properties to the neural networks. Thus it is considered as one of the essential ingredients among other building blocks of a neural network. But the selection of the appropriate activation function for the enhancement of model accuracy is strenuous in a sense; the performance of the NN-model is influenced by a proper selection of activation function for a dataset. Proper activation function selection is still a trial and error method for which the model accuracy improves for classification. As a solution to this problem, we have proposed an activation function ensembling by majority voting that has significantly improved the model accuracy in a classification context. The proposed model is tested on four benchmark datasets such as MNIST, Fashion MNIST, Semeion, and ARDIS IV datasets. The result shows that the performance of the proposed model is appreciably better than other traditional methods such as Convolutional Neural Network (CNN), Support Vector Machine (SVM), Recurrent Neural Network (RNN).

Protein structure prediction (PSP) from its primary sequence is a challenging task in computational biology. PSP is an optimization problem that determines the stable or native structure with minimum free energy. Several researchers have applied various heuristic algorithms and/or their variants to solve this problem. However, the mechanism to select a particular algorithm is not known a priori. Fitness landscape analysis (FLA) is a technique to determine the characteristics of a problem or its structural features based on which the most appropriate algorithm can be recommended for solving the problem. The aim of this study is two-fold while considering the PSP problem. Firstly, the structural features are determined by using the standard FLA techniques and secondly, the performance of some of the well-known optimization algorithms are analyzed based on the structural features as an illustration of the usefulness of the former research agenda. In this paper, we determine structural features of the PSP problem by analyzing the landscapes generated by using the quasi-random sampling technique and city block distance. Comprehensive simulations are carried out on both artificial and real protein sequences in 2D and 3D AB off-lattice model. Numerical results indicate that the complexity of the PSP problem increases with protein sequence length. We calculate the Pearson correlation coefficient between the FLA measures, separately for 2D and 3D off-lattice model and significant differences are identified among the measures. Six well-known real-coded optimization algorithms are evaluated over the same set of protein sequences and the performances are subsequently analyzed based on the structural features. Finally, we suggest the most appropriate algorithms for solving different classes of PSP problem.

Drug is a very much essential substance in health care system. Producing a new drug for a disease in the market using traditional method is very time consuming and expensive. Recently, drug design process is sped-up by using computer resources known as Computer-Aided Drug Design (CADD). In drug design, bio-molecules are responsible to produce a new drug. Therefore, molecules identification is an essential part of CADD. In this paper, two machine learning algorithms such as Naive Bayesian (NB) classifier and k-Nearest Neighbors (k-NN) are evaluated to classify multi-class drug molecules. The result reveals that k-NN method shows higher Accuracy and higher Precision compare to NB. Furthermore, Recall and F1-score of k-NN are higher than that of NB.

Traditional drug designing and discovery processes are a very time consuming, expensive and challenging tasks to produce a new drug in the market. Computer Aided Drug Design (CADD) is a promising approach that is cost effective as well as speeds up the drug designing process. CADD is a computational methods which provides resources for simplifying the design and discovery of a new drug. At molecular level, a drug binds to the target protein and neutralizes the disease. Therefore, identification of active molecules which can bind to the target protein is an essential part of CADD. In this paper, the back propagation neural network model is employed for predicting inactive or active molecules, which provides chemical compounds with desirable properties for drug design. The proposed approach demonstrates 99% prediction accuracy on the dataset which consists of active and inactive molecules taken from PubChem data repository.

Artificial Neural Networks (ANNs) have been successfully applied to solve various classification problems. To perform the classification task,neural network training is essential. Multiple parameters are associated with neural network training and setting these parameters are a very challenging task. Several meta-heuristic algorithms are employed to train the network. In this paper, Particle Swarm optimization (PSO) is used to train the network to optimize the training error while solving a classification problem. Performance of the proposed method is evaluated on ten different data sets which are in supervised. The experimental result shows a significant performance of the proposed methodology compared to traditional neural network training approaches.

Extensive research on heuristic algorithms has proved their potential in solving complex optimization problems. However, it is not easy to choose the best heuristic technique for solving a particular problem. Fitness landscape analysis is used for understanding the problem characteristics based on which the best-suited algorithm for the problem can be chosen. Compared to the literature on discrete search spaces, only a few significant works have been undertaken on landscape analysis in continuous search spaces. Random walk (RW) algorithm has been used for generating sample points in the search space, and fitness landscape is created based on the relative fitness of the neighboring sample points. This paper proposes a chaos-based random walk algorithm, called as the chaotic random walk (CRW), applied in continuous search space to generate the landscape structure for a problem. The chaotic map is used to generate the chaotic pseudorandom numbers for determining variable scaled step size and direction of the proposed RW algorithm. Histogram analysis demonstrates better coverage of search space by the CRW algorithm compared to the simple and progressive random walk algorithms. In addition, we test the efficiency of the proposed method by quantifying the ruggedness and deception of a problem using entropy and fitness distance correlation measures. Experiments are conducted on the IEEE Congers on Evolutionary Computing 2013 benchmark functions in continuous search space having different levels of complexity. Extensive experiments indicate the capability for generating landscape structure on the continuous search space and efficiency of the proposed method to investigate the structural features of fitness landscapes.

Stochastic optimization algorithms have potential to solve optimization problems in various fields of engineering and science. However, increasing non-linearity, non convexity, multi-modality, discontinuity, and even dynamics make the problems more complex and intractable. Classical optimization techniques are not able to determine global solution by analyzing rough non-linear surfaces. Heuristic algorithms have been used for determining global solution for this type of problems. However, heuristic algorithm is knowledge dependent, so finding a unique heuristic optimization algorithm for obtaining optimum solutions for all problems is not feasible. Hybridization is an integrated framework where merits of algorithms are utilized to improve performance of the optimizers. Particle Swarm Optimization (PSO) and Differential Evolution (DE) algorithm are two heuristic algorithms despite certain shortcomings have been applied to solve global optimization problems. In this paper, we propose an integrated framework...

This chapter provides a comprehensive overview of the protein structure prediction problem based on metaheuristic algorithms. At first, the basic concepts of proteins, the level of protein structure have been presented in a formal way. A computational model, as well as techniques, have been addressed for solving protein structure prediction (PSP) problem. The chapter discusses the basic fundamentals of metaheuristics algorithms in detail and finally ends with a discussion of techniques are used in the book towards solving the problem.

This chapter introduces four schemes for protein structure prediction (PSP) based on 2D and 3D AB off-lattice model. The proposed methods are based on the modified versions of the classical PSO, BA, BBO and HS algorithms, providing an improved solution using different strategies. The strategies are developed in order to find global minimum energy value over the multi-modal landscape structure of the PSP problem. The performance of the proposed methods are extensively compared with the algorithms which are applied to the PSP problem over a test suite of several artificial and real-life protein instances.

This chapter describes a chaos based random walk (CRW) algorithm for analyzing landscape structure in continuous search spaces. Unlike the existing random walks, no fixed step size is required in the proposed algorithm, rather conduct the random walk. The chaotic map is used to generate the chaotic pseudo random numbers (CPRN) for determining the variable-scaled step size and direction. The superiority of the new method has been demonstrated while comparing it with the simple and progressive random walk algorithms using histogram analysis. The performance of the proposed CRW algorithm is evaluated on the IEEE Congers on Evolutionary Computation (CEC) 2013 benchmark functions in continuous search space having different levels of complexity. The proposed method is applied to analyzing the landscape structure for protein structure prediction problem in continuous search space.

Hybridization is an integrated framework that combines the merits of algorithms to improve the performance of an optimizer. In this chapter, the synergism of the improved version of particle swarm optimization (PSO) and differential evolution (DE) algorithms are invoked to construct a hybrid algorithm. The proposed method is executed in an interleaved fashion for balancing exploration and exploitation dilemma in the evolution process. The results are tested on ten real protein instances, taken from the protein data bank. The effectiveness of the proposed algorithm is evaluated through qualitative and quantitative comparisons with other hybridization of PSO and DE; and comprehensive learning PSO algorithms.

This chapter provides an overview of the research in protein structure prediction with metaheuristic techniques using AB off-lattice model. The chapter discuses related work covered under the heads of metaheuristic methods, classified into four categories. The scope of work has been outlined and finally, the chapter ends with a discussion of the contributions we have made in this book.

Fitness landscape analysis (FLA) is a technique to determine the characteristics of a problem or its structural features based on which the most appropriate algorithm is possible to recommend for solving the problem. In this chapter, we determine structural features of the protein structure prediction problem by analyzing the landscape structure. A landscape of the protein instances is generated by using the quasi-random sampling technique and city block distance. Structural features of the PSP Landscape are determined by applying various landscape measures. Numerical results indicate that the complexity of the PSP problem increases with protein sequence length. Six well-known real-coded optimization algorithms are evaluated over the same set of protein sequences and the performances are subsequently analyzed based on the structural features. Finally, we suggest the most appropriate algorithm(s) for solving the PSP problem.

Differential evolution (DE) algorithm is a population based stochastic search technique widely applied in scientific and engineering fields for global optimization over real parameter space. The performance of DE algorithm highly depends on the selection of values of the associated control parameters. Therefore, finding suitable values of control parameters is a challenging task and researchers have already proposed several adaptive and self-adaptive variants of DE. In the paper control parameters are adapted by levy distribution, named as Levy distributed DE (LdDE) which efficiently handles exploration and exploitation dilemma in the search space. In order to assure a fair comparison with existing parameter controlled DE algorithms, we apply the proposed method on number of well-known unimodal, basic and expanded multimodal and hybrid composite benchmark optimization functions having different dimensions. The empirical study shows that the proposed LdDE algorithm exhibits an overall better performance in terms of accuracy and convergence speed compared to five prominent adaptive DE algorithms.

Protein structure prediction (PSP) from its amino acid sequence is a challenging problem in computational biology and can be considered as a global optimization problem. It is a multi-modal optimization problem and belongs to NP-hard class. In this paper, Biogeography Based Optimization with Chaotic Mutation (BBO-CM) algorithm has been developed to optimize 3D protein structure. The proposed algorithm prevents premature convergence and jumping out from the local minima during execution and converges with the optimum solution. Chaos system generates the chaotic pseudo random sequence which is utilized in mutation operation of BBO algorithm to increase the population diversity. The experiments are carried out with artificial and real protein sequences with different length to confirm the performance and robustness of the BBO-CM algorithm. Results are compared with other algorithms demonstrating the efficiency of the proposed approach.

Protein structure prediction (PSP) is an important research area in bio-informatics for its immense scope of application in drug design, disease prediction, name a few. Structure prediction of protein based on sequence of amino acids is a NP-hard and multi-modal optimization problem. This paper presents an improved harmony search (ImHS) algorithm to solve the PSP problem based on 3D off lattice model. In the proposed method, the basic harmony search (HS) algorithm combined with dimensional mean based perturbation strategy to avoid premature convergence and enhance the capability of jumping out from the local optima. The experiments are carried out on a set of real protein sequences with different length collected from the Protein Data Bank (PDB) to validate the efficiency of the proposed method. Numerical results show that the ImHS algorithm significantly outperforms compared to other algorithms on protein energy minimization.

In the field of Image Processing, Image segmentation is a low level but important task in entire image understanding system which divides an image into its multiple disjoint regions based on homogeneity. In most of the machine vesion and high level image understanding application this is one of the important steps. Till date different techniques of image segmentation are available and hence There exists a huge survey literature in different approaches of Image Segmentation. Selection of image segmentation technique is highly problem specific. There is no versatile algorithm which is applicable for all kinds of images. Optimization based image segmentation is not explored much which can be applied to reduce complexity of the problem. The aim of the paper is to search for an optimized threshold value for Image Segmentation using Particle Swarm Optimization (PSO) algorithm where fitness function is designed based on entropy of the image.

In recent decade Evolutionary Algorithms plays an important role in many engineering design and optimization problems. Particle Swarm Optimization (PSO) is one of such algorithm which is based on the intelligent food searching behavior of swarm like birds flock, fish schooling. It has been shown that it works efficiently on noisy, multimodal and composite functions. However, it stuck at local optima at later stage of evolution due to unexplore the search space. Several variations of pso and mutation based approached was developed for this problem. In this paper, an adaptive mutation is proposed for multiobjective pso and called it AMPSO. In AMPSO, mutation is applied on the position and velocity of the particles based on the fitness values of the particles. Proposed algorithm carried on 5 multiobjective benchmark functions. The experimental results shown the better performance comparing with other algorithms in terms of best, mean and standard deviation.

Protein Structure Prediction (PSP) using sequence of amino acids is a multimodal optimization problem and belongs to NP hard class. Researchers and scientists put their efforts to design efficient computational intelligent algorithm for solving this kind of problem. Bees Algorithm (BA) is a swarm intelligence based algorithm inspired by the foraging behaviour of honey bees colony, already exhibits its potential ability for solving optimization problems. However, it may produce premature convergence when solving PSP like problems. To prevent this situation, Adaptive Polynomial Mutation based Bees Algorithm (APM-BA) has been proposed in this paper for predicting protein structure in 2D AB off-lattice model. In this strategy, each of best scout bees are mutated with adaptive polynomial mutation technique when their performances are no more improve during execution phase. The experiments are conducted on artificial and real protein sequences and numerical results show that the proposed algorithm has strong ability for solving PSP problem having minimum energy.

Large-scale global optimization is one of the most challenging problems in the domain of stochastic optimization. Due to high dimensionality in the entire optimization process, different types of problems may occur for finding the global optima, e.g., solution space increases exponentially, problem complexity increases, and candidate search direction also increases exponentially. So, deterministic optimization algorithms cannot perform well for this kind of problems. Differential evolutionary algorithm is a population-based, stochastic search and optimization algorithm which can be used for global optimization problems. In this paper, we present self-adaptive dynamic population-based differential evolutionary algorithm which automatically adapts its parameters including population size.

Image segmentation is the process of subdividing a digital image into its constitute regions such that pixels belong to the same region will be same based on some image property (such as grayscale value, color, texture) and pixels in the different group will be different based on the same image property. Till date, different researchers have taken image segmentation problem from a different point of view and developed several image segmentation algorithms. This paper is going to address an optimization-based approach in color image segmentation where optimized threshold value is chosen by maximizing the Kapur’s Entropy Function.

The Particle Swarm Optimization (PSO) algorithm is a novel population based swarm algorithm has shown good performance on well-known numerical test problems. However, PSO tends to suffer from premature convergence on multimodal test problems. This is due to lack of diversity of population in search space and leads to stuck at local optima and ultimately fitness stagnation of the population. To enhance the performance of PSO algorithms, in this paper, we propose a method of population adaptation (PA). The proposed method can identify the moment when the population diversity is poor or the population stagnates by measuring the Euclidean distance between particle position and particles average position of a population. When stagnation in the population is identified, the population will be regenerated by normal distribution to increase diversity in the population. The population adaptation is incorporated into the PSO algorithm and is tested on a set of 13 scalable CEC05 benchmark functions. The results show that the proposed population adaptation algorithm can significantly improve the performance of the PSO algorithm with standard PSO, ATREPSO and ARPSO.

In this paper, particle swarm optimization (PSO) with levy flight is proposed. PSO is a population based global optimization algorithm has faster convergence but often gets stuck in local optima due to lack of diversity in the population. In the proposed method, levy flight is applied on a percentage of particles excluding global best particle to create diversity in population. Adaptive polynomial mutation is applied on global best (gbest) particle to get it out from the trap in local optima. The method is applied on well-known benchmark unconstrained functions and results are compares with classical PSO. Form the experimental result, it has been observed that the proposed method performs better than classical PSO.

Artificial Bee Colony (ABC) is a swarm based stochastic search algorithm inspired by the foraging behavior of honeybees. Due to the simplicity of implementation and promising optimization capability, ABC is successfully applied to solve wide class of scientific and engineering optimization problems. But, it has problems of premature convergence and trapping in local optima. In this paper, to enhance the performance of ABC, we have proposed a modified version of ABC algorithm using Differential Evolution (DE) and Polynomial Mutation (PM) called DE-PM-ABC. The comparison with ABC by Karaboga [1], MABC [27] by Liu et al. using some benchmark functions of CEC 2005 demonstrates that our approach achieves a good trade-off between exploration and exploitation and thus obtains better global optimization result and faster convergence speed.

Particle Swarm Optimization (PSO) algorithm is a swarm based algorithm deliver good performance in many optimization problems. However, PSO has tendency of trapping into local optima. In the paper, an improved PSO algorithm has been proposed by employing Exploratory Move on global best particle of the swarm called as PSO with exploratory move (ExPSO) algorithm. In the proposed approach in order to preventing PSO algorithm from trapping into local optima, particles are jumped to an unknown position made by the exploratory move. The performance of the ExPSO algorithm has been investigated on a set of eight standard benchmark functions and results are compared with the simple PSO, constriction factor PSO (CFPSO) and inertia weight PSO (IWPSO). The numerical results show that the ExPSO algorithm performs better, robust and statistically significant on most of the test cases.

Differential Evolution (DE) algorithm is a real parameter encoded evolutionary algorithm for global optimization. In this paper, Levy distributed DE (LevyDE) has been proposed. The main objective of LevyDE algorithm is to introduce a parameter control mechanism in DE based on levy distribution, a heavy tail distribution, for both the mutation and crossover operations. The main emphasis of this paper is to analyze the behavior and dynamics of the LevyDE and make a comparison with other standard algorithms such as DE/best/1/bin [1], DE/rand/1/bin [1] and ACDE [8] on basis of CEC’05 benchmark functions.

Protein Structure Prediction with lowest energy from its primary sequence of amino acids is a complex and challenging problem in computational biology, addressed by researchers using heuristic optimization techniques. Particle Swarm Optimization (PSO), a heuristic optimization technique having strong global search capability but often stuck at local optima while solving complex optimization problem. To prevent local optima problem, PSO with local search (HPSOLS) capability has been proposed in the paper to predict structure of protein using 2D off-lattice model. HPSOLS is applied on artificial and real protein sequences to conform the performance and robustness for solving protein structure prediction having lowest energy. Results are compared with other algorithms demonstrating efficiency of the proposed model.

In this paper, a new version of Particle Swarm Optimization (PSO) Algorithm has been proposed where the velocity update equation of PSO has been modified. A new term is added withthe original velocity update equation by calculating difference between the global best of swarm and local best of particles. The proposed method is applied on eight well known benchmark problems and experimental results are compared with the standard PSO (SPSO). From the experimental results, it has been observed that the newly proposed PSO algorithm outperforms the SPSO in terms of convergence, speed and quality.

Several population based search algorithms are developed by the researchers to predict the native state of protein from its primary sequences. The paper aims at predicting the native conformation of proteins in lattice model using PSO based searching method. However, stuck at local minima and generating illegal conformation are the main drawbacks of applying the search algorithm in protein structure prediction. Adaptive Polynomial Mutation (APM) is performed to remove local minima while illegal conformations are repaired using backtracking method. Benchmark sequences with different length are applied to verify the proposed algorithm showing better results compare to the earlier approaches.

In this work, we proposed a method of artificial neural network learning using differential evolutionary(DE) algorithm. DE with global and local neighborhood based mutation(DEGL) algorithm is used to search the synaptic weight coefficients of neural network and to minimize the learning error in the error surface.DEGL is a version of DE algorithm in which both global and local neighborhood-based mutation operator is combined to create donor vector.The proposed method is applied for classification of real-world data and experimental results show the efficiency and effectiveness of the proposed method and also a comparative study has been made with classical DE algorithm.

Protein Structure Prediction (PSP) is a challenging problem in bioinformatics and computational biology research for its immense scope of application in drug design, disease prediction, name a few. Developing a suitable optimization technique for predicting the structure of proteins has been addressed in the paper, using Differential Evolutionary (DE) algorithm applied in the square 2D HP lattice model. In the work, we concentrate on handling infeasible solutions and modify control parameters like population size (NP), scale factor (F), crossover ratio (CR) and mutation strategy of the DE algorithm to improve its performance in PSP problem. The proposed method is compared with the existing methods using benchmark sequence of protein databases, showing very promising and effective performance in PSP problem.

In this work, a method to control the parameters of Differential Evolution (DE) algorithm is proposed. Here the control parameters of DE are co-evolved by Particle Swarm Optimization (PSO) algorithm. The classical DE algorithm has two main control parameters: Scale Factor (F) and Cross-over Rate (CR). These are selected on trial-and-error basis for solving optimization problems. Several optimization problems lead to optimal or sub-optimal solution by proper selection of control parameters of the DE algorithm. In this proposed method, PSO algorithm is used to tune the scale factor and cross-over rate in DE algorithm. Basically PSO algorithm is used as a meta-optimizer for DE algorithm. The proposed method is termed as mPSO-DE in this paper. The mPSO-DE algorithm is applied on 12 benchmark unconstrained optimization problems. The obtained results are compared with that of classical DE algorithm. From the experimental studies, it has been found that the proposed mPSO-DE algorithm performed better than DE algorithm.

Particle Swarm Optimization (PSO) has shown its good search ability in many optimization problem. But PSO easily gets trapped into local optima while dealing with complex problems. In this work, we proposed an improved PSO, namely PSO-APM, in which adaptive polynomial mutation strategy is employed on global best particle with the hope that it will help the particles jump out local optima. In this work, we carried out our experiments on 8 well-known benchmark problems. Finally the results are compared with classical PSO and PSO with power mutation (PMPSO).

Constrained function optimization using particle swarm optimization (PSO) with polynomial mutation is proposed in this work. In this method non-stationary penalty function approach is adopted and polynomial mutation is performed on global best solution in PSO. The proposed method is applied on 6 benchmark problems and obtained results are compared with the results obtained from basic PSO. The experimental results show the efficiency and effectiveness of the method.

Particle swarm optimisation PSO is population-based optimisation algorithm having stochastic in nature. PSO has quick convergence speed but often gets stuck into local optima due to lacks of diversity. In this work, first mutation operator adopted from Differential Evolution DE algorithm is applied in PSO with decreasing inertia weight PSO-DMLB. In second method, DE mutation is applied in another PSO variant, namely Comprehensive Learning PSO CLPSO. The second method is termed as CLPSO-DMLB. Local best position of each particle is muted by a predefined mutation probability with the scaled difference of two randomly selected particle's local best position to increase the diversity in the population to achieve better quality of solutions. The proposed methods are applied on well-known benchmark unconstrained functions and obtained results are compared to show the effectiveness of the proposed methods.