Search

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).

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.

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.