** GRADIENT-DESCENT**(
)

Each training example is a pair of the form, whereis the vector of input values, andtis the target output value.is the learning rate (e.g., .05).

- Initialize each
*w*_{i}to some small random value - Until the termination condition is met, Do
- Initialize each to zero.
- For each
in
,
Do
- Input the instance
to the unit and compute the output
*o* - For each linear unit weight
*w*_{i}, Do

- Input the instance
to the unit and compute the output
- For each linear unit weight
*w*_{i}, Do