Machine learning
Writing software used to mean that its human creator had to predict and outline all possible inputs and each respective output; if this happens, do that. Instead, machinelearning software analyses examples or data sets and infers how to react to an unforeseen input. Rather than explicitly instructing the software, “If you see a photo with pointy ears, fur, a tail, paws, identify it as a cat,” a machine learning system looks at photos of cats. It identifies the common distinguishing features (pointy ears, fur, tail), uses those criteria to look at a photo that it has never seen before, and concludes that, yes, that is a cat.
Artificial neural network
A neural net is a way to organise machine learning software, so named because the software’s nodes, or…
