作者: Ralf Herbrich , Max Keilbach , Thore Graepel , Peter Bollmann-Sdorra , Klaus Obermayer
DOI: 10.1007/978-1-4615-5029-7_7
关键词:
摘要: Neural Networks – originally inspired from Neuroscience provide powerful models for statistical data analysis. Their most prominent feature is their ability to “learn” dependencies based on a finite number of observations. In the context term “learning” means that knowledge acquired samples can be generalized as yet unseen this sense, Network often called Learning Machine. As such, might considered metaphor an agent who learns his environment and thus infers strategies behavior limited contribution, however, we want abstract biological origins rather present them purely mathematical model.