In this paper an inclined planes optimization algorithm, is used to optimize the performance of the multilayer perceptron. Indeed, the performance of the neural network depends on its parameters such as the number of ...
详细信息
ISBN:
(纸本)9781509064540
In this paper an inclined planes optimization algorithm, is used to optimize the performance of the multilayer perceptron. Indeed, the performance of the neural network depends on its parameters such as the number of neurons in the hidden layer and the connection weights. So far, most research has been done in the field of training the neural network. In this paper, a new algorithmoptimization is presented in optimal architecture for data classification. Neural network training is done by backpropagation (BP) algorithm and optimization the architecture of neural network is considered as independent variables in the algorithm. The results in three classification problems have shown that a neural network resulting from these methods have low complexity and high accuracy when compared with results of Particle Swarm optimization and Gravitational Search algorithm.
In real world optimization problems, sometimes not just a single solution but a collection of multiple solutions which are scattered throughout the search space are to be found. Over the years heuristic algorithms hav...
详细信息
ISBN:
(纸本)9781728186290
In real world optimization problems, sometimes not just a single solution but a collection of multiple solutions which are scattered throughout the search space are to be found. Over the years heuristic algorithms have come to be recognized as one of the leading methods in solving such problems, also known as multimodal optimization problems. In this paper, one of the recent heuristic methods called inclinedplanesoptimization (IPO) is employed to solve multimodal optimization problems. To prove the efficiency of the proposed method, six standard benchmark functions are used. Furthermore, the proposed method is compared with other similar methods such as genetic algorithm (GA) and particle swarm optimization (PSO). Experimental results indicate that IPO has a superior performance compared to similar methods.
暂无评论