Artificial neural networks (ANNs) are adaptable systems that can solve problems that are difficult to describe with a mathematical relationship. They seek relationships between different types of datasets with their abilities to learn either with supervision or without. ANNs recognize patterns between input and output space and generalize solutions, in a way simulating the human brain’s learning experience with many relatively simple individual processing elements, called neurons. Neurons are networked (network topology) in a number of ways depending on the problem type and complexity. One of the most widely used ANN learning techniques is supervised learning coupled with a multilayer perceptron (MLP) topology due to its flexible applicability to a wide range of modeling problems involving both general classification and regression. ANNs, due to this flexibility, have been applied to many fields since the 1990s and their theory, types (such as radial basis functions, random...
|Title||Multilayer perceptrons (MLPs)|
|Authors||C. Özgen Karacan|
|Publication Type||Book Chapter|
|Publication Subtype||Book Chapter|
|Record Source||USGS Publications Warehouse|
|USGS Organization||Eastern Energy Resources Science Center; Geology, Energy & Minerals Science Center|