Python Deep Learning - Second Edition

Linear regression

We have already introduced linear regression in Chapter 1, Machine Learning – an Introduction. To recap, regarding utilization of the vector notation, the output of a linear regression algorithm is a single value, y , and is equal to the dot product of the input values x and the weights w: . As we now know, linear regression is a special case of a neural network; that is, it's a single neuron with the identity activation function. In this section, we'll learn how to train linear regression with gradient descent and, in the following sections, we'll extend it to training more complex models. You can see how the gradient descent ...