Deep generative models are an important term in the fields of artificial intelligence, big data and smart data, as well as digital transformation. They describe a special type of artificial intelligence that is capable of „inventing“ new data. This can be, for example, images, texts, voices or even pieces of music that look or sound as if they were made by a human.
The special thing about deep generative models is that they are fed with a lot of sample data, for example, thousands of photos, and learn to recognise typical patterns from them. This creates a kind of „creativity“ with which the model can create new, completely unique content.
A striking example: using a deep generative model, designers can create a multitude of new, realistic-looking product images from just a few photos of furniture. This saves enormous costs on photoshoots and enables rapid development of new design ideas.
Deep generative models are revolutionising many industries through their ability to create entirely new content. For example, they are used in marketing to create personalised advertising images or help companies gain new insights from large amounts of data.
Deutsch 
English (UK)