作者: Teijiro Isokawa , Haruhiko Nishimura , Nobuyuki Matsui
DOI: 10.1109/IJCNN.2012.6252536
关键词:
摘要: Choosing an appropriate activation function is a challenging problem in quaternionic neural networks, due to the analyticity domain. This paper presents Hopfield-type network with equivalent of tanh function. obtained by incorporating so-called “local analyticity” on domain and recent results complex-valued functions. The stability shown proving monotonic decrease energy changes neuron states.