Some new thoughts on Neural Networks with complex connection matrices
Miklos N. Szilagyi, Jens Christian Mikkelsen, Kjeld Høyer Mortensen · DAIMI Report Series · 1990
In the literature neural networks are usually understood in the domain of real activation functions and connection weights. The concept of neural networks was generalized by M. Szilagyi to include complex connections between complex units. In the present note this idea is further developed. A mathematical model is presented, an expression for the network's energy as well as a complex learning rule are proposed.