Analysis of A Mixed Neural Network Based on CNN and RNN for Computational Model of Sensory Cortex

Abstract

Under the limitation of modern science, Anatomy is not able to discover the extremely complicated organ - brain’s working activities. Along with the development of machine learning and its subset - neural network, named by nearly implementing neurons’ connection, scientists found an efficient way to represent brain as a visible approach. Therefore, a new discipline has been created under biology, named as computational neuroscience. Most scientists focus on finding computational models that are closed to fit the specific or general areas of the cortex. Convolutional neural network (CNN) and Recurrent neural network (RNN) are two of them. By taking advantages of those two most popular networks, in this study, a mixed computational model with CNN and RNN might be the best computational model for analogizing brain’s activities so far.