A LSTM based deep learning network for recognizing emotions using wireless brainwave driven system

Abstract

Asynchronous differential evolution (ADE) supports parallel optimization and effective exploration. The updation in population is done immediately when a vector with better fitness is found in ADE algorithm. The working of ADE and Differential Evolution (DE) is similar except the instant population updation feature and asynchronous nature. In this paper, we have integrated ADE with successful parent-selecting (SPS) framework and trigonometric mutation to enhance the performance. Additionally, the control parameters are updated in an adaptive manner to support better exploration as well as exploitation. The proposed algorithm is named as SPS embedded adaptive ADE with trigonometric mutation (SPS-AADE-TM). The modified mutation operation and adaptive parameters can increase the population diversity and the convergence speed. The parameter adaptation feature can automatically obtain the appropriate values of control parameters to enhance the robustness of SPS-AADE-TM. The proposed algorithm is tested over twenty-five widely used bench-mark functions and four engineering design problems. Two nonparametric statistical tests are also carried out to validate the performance of SPS-AADE-TM. The simulation results show that the proposed work provides promising results and outperforms the competitive algorithms.