The initial values of weights for each input are transformed using the equation shown earlier to assign the values to the next input (i.e., row). The threshold values (0.5) sets the output Y to 1 in the next row if the weighted sum of inputs is greater than 0.5; otherwise, Y is set to 0. In the first step, two of the four outputs are incorrect (delta = 1), and a consistent set of weights has been found. In subsequent steps, the learning algorithm improves the results until if finally produces a set of weights that give the correct results (W1 = W2 = 0.7 in step 4 of Table 6.1). Once determined, a neuron with these weights can quickly perform the OR operation.