Deep belief networks (DBN) are at home in the fields of artificial intelligence, big data, smart data and digital transformation. They are among the modern methods with which computers and machines learn to recognise patterns and correlations in large amounts of data.
Imagine Deep Belief Networks as a network of several layers connected in series, similar to the layers of a cake. Each layer takes care of a specific detail, for example the recognition of colours, edges or shapes on an image. The deeper the layers, the more complex the patterns that can be recognised.
An illustrative example: a deep belief network can teach a computer to reliably recognise handwritten numbers on a form - even if each person writes something different. To do this, the network analyses many different images of numbers, learns from these examples and ultimately recognises new, unknown numbers.
Deep Belief Networks (DBN) are used wherever large amounts of data need to be processed and understood automatically, for example in image or speech recognition or when analysing customer data.
Deutsch 
English (UK)