Cross-domain transfer learning is a term used in the fields of artificial intelligence, big data, smart data and digital transformation. It describes a method in which knowledge from one application area (domain) is transferred to another. This may sound complicated at first, but it is very practical: artificial intelligence does not have to start from scratch, but can use what it has already learnt from another sector.
Imagine an AI that has learned to recognise tumours on X-ray images in medicine. This knowledge can be used across domains to detect defects on industrial X-ray images in production - without any completely new training. The AI utilises the skills it has already acquired from one domain and "transfers" them to the other.
Cross-domain transfer learning saves time, costs and computing power. This enables companies to bring innovations to different areas more quickly as they can utilise existing expertise more intelligently. This is particularly exciting for companies that want to efficiently utilise diverse data from different sources in order to make processes more intelligent and flexible.
Deutsch 
English (UK)