다양한 활성 함수를 사용하는 신경회로망의 구성

Neural Networks with Mixed Activation Functions

When we apply the neural networks to applications, we need to select proper architecture of the network and the activation function of the network is one of most important characteristics. In this research, we propose a method to make a network using multiple activation functions. The performance of the proposed method is investigated through the computer simulations on various regression problems.