Abstract
Machine learning (ML) is any predictive algorithm, or a combination of algorithms, that learns from data (that learns from "experience"), and makes predictions without being explicitly coded with a deterministic model. The most immediate example are neural networks, which are trained with data to minimise a cost function and perform predictions. In this work, we present some ML methods, with simple examples to grasp the basic concepts of each algorithm, avoiding formal mathematical complexities. The techniques involved in ML include gradient methods, genetic algorithms, simulated annealing, neural networks, and the novel field of quantum computing as an aid to speed the algorithms. Geophysical examples are given to illustrate practical applications.