Representation learning

Representation learning is a way of building machines (like computers or robots) that can understand the world. It helps to think of this like learning a language. When we learn a language, we first learn the individual words, then we learn how to put them together to make sentences (or representations). Representation learning is kind of like this - but for machines. Machines learn how to take the pieces of information they are presented with (like words, images, or sounds) and put them together in a way that makes it easier for them to understand the world and take action.