Abstract: In this paper, an intelligent system using an artificial neural network uses a genetic algorithm as a training method. This hybrid of an artificial neural network (ANN) and a genetic algorithm (GA) results in better solutions being derived in less time than would be derived from training an ANN using the back-propagation algorithm. The new algorithm outperforms the traditional method by a significant amount. The hybrid genetic algorithm is designed and experimented against the back propagation and the number of operations to reach convergence is reduced from 4,000 epochs to less than 50 generations, resulting in a 51.2% decrease in training time. These results indicate that the GA approach is a useful tool for training an ANN.
https://dl.acm.org/doi/10.1145/1655925.1662625
Abstract: In this paper, an intelligent system using an artificial neural network uses a genetic algorithm as a training method. This hybrid of an artificial neural network (ANN) and a genetic algorithm (GA) results in better solutions being derived in less time than would be derived from training an ANN using the back-propagation algorithm. The new algorithm outperforms the traditional method by a significant amount. The hybrid genetic algorithm is designed and experimented against the back propagation and the number of operations to reach convergence is reduced from 4,000 epochs to less than 50 generations, resulting in a 51.2% decrease in training time. These results indicate that the GA approach is a useful tool for training an ANN. https://dl.acm.org/doi/10.1145/1655925.1662625