Biased Neuron is for neuron's with a bias, e.g.
An interface for updates rules that make use of an upper and lower bound, either for clipping or for setting intrinsic bounds.
Interface for update rules which can clip their values above and below upper and lower bounds (clipping can be turned on or off).
Indicates that an update rule is differentiable, and has a getDerivative function.
Indicates that an update rule is invertible.
An interface which should be implemented by any neuron update rule that uses a randomizer to provide noise.
The network.neuron_update_rules.interfaces package contains abstract classes and interfaces relating to neuron update.