Dropout Probability Estimation in Convolutional Neural Networks by the Enhanced Bat Algorithm

摘要: In recent years, deep learning has reached exceptional accomplishment in diverse applications, such as visual and speech recognition, natural language processing. The convolutional neural network represents a particular type of commonly used for the task digital image classification. A common issue models is high variance problem, or also called over-fitting. Over-fitting occurs when model fits well with training data fails to generalize on new data. To prevent over-fitting, several regularization methods can be used; one powerful method dropout regularization. find optimal value rate very time-consuming process; hence, we propose by utilizing metaheuristic algorithm instead manual search. this paper, hybridized bat probability compare results similar techniques. experimental show that proposed hybrid overperforms other