Hands-On Artificial Intelligence for Cybersecurity

Introducing the gradient

In mathematical terms, a gradient represents the partial derivative that's calculated on a given point in the n-dimensional space; it also represents the tangent line (slope) of the point that's being considered.

The gradient is used in machine learning as a cost function to be minimized in order to reduce the prediction errors that are produced by the algorithms. This consists of minimizing the difference between the value estimated by the algorithm and the observed value.

The minimization method that's used is known as gradient descent, which is a method of optimizing the combination of weights to be assigned to the input data in order to obtain the minimum difference between the values estimated and the values ...