Winner-take-all circuits

特許 係属中 Winner-take-all is a computational principle applied in computational models of neural networks by which neurons in a layer compete with each other for activation. In the classical form, only the neuron with the highest activation stays active while all other neurons shut down, however other variations that allow more than one neuron to be active do exist, for example the soft winner take-all, by which a power function is applied to the neurons.