Correlation Learning Rule | Artificial Neural Networks
It is a supervised algorithm learning rule.
This rule follows a principle similar to the hebbian learning rule, where;
At the start, values of all weights connecting neurons are set to zero, and it’s assumed that, neurons that are either positive or negative at the same time have strong positive weights while those that are opposite, positive and negative, have strong negative weights.
This rule, however, unlike the hebbian learning rule, the desired response is used to calculate the change in weights for the purpose of improving them, making it a supervised learning algorithm learning rule.
Correlation Learning Rule | ANN Learning Rules
Artificial Neural Networks | thetqweb