In artificial intelligence, the delta rule is a gradient descent learning rule for updating the weights of the artificial neurons in a single-layer perceptron. It is a special case of the more general backpropagation algorithm. For a neuron with activation function, the delta rule for 's th weight is given by
- ,
where
| is a small constant called learning rate | |
| is the neuron's activation function | |
| is the target output | |
| is the weighted sum of the neuron's inputs | |
| is the actual output | |
| is the th input. |
It holds that and .
The delta rule is commonly stated in simplified form for a perceptron with a linear activation function as
Read more about Delta Rule: Derivation of The Delta Rule
Famous quotes containing the word rule:
“The earliest instinct of the child, and the ripest experience of age, unite in affirming simplicity to be the truest and profoundest part for man. Likewise this simplicity is so universal and all-containing as a rule for human life, that the subtlest bad man, and the purest good man, as well as the profoundest wise man, do all alike present it on that side which they socially turn to the inquisitive and unscrupulous world.”
—Herman Melville (1819–1891)