Gradient Type Learning Rules for Neural Networks Based on Watcher-Environment Model

Moderatism, which is a learning rule for ANNs, is based on the principle that individual neurons and neural nets as a whole try to sustain a “moderate” level in their input and output signals. In this way, neural network receives feedback signal from the outside environment, and the principle of connectionism is preserved. In this paper, a potential Moderatism-based local, gradient learning rule based on a watcher-environment model is proposed.