Dynamical properties of neural nets using competitive activation mechanisms
Michel Benaı̈m, Manuel Samuelides · 1990
It is shown that layered feedforward nets are convergent, while nonperiodic oscillations might occur in bidirectional nets. The virtual lateral inhibition phenomenon in feedforward nets is discussed. It is shown that a certain kind of nonlinearity in output functions may be successfully used in realizing winner-take-all networks