Few-shot learning is a term from the field of artificial intelligence and digital transformation. It describes a method in which an artificial intelligence learns to solve new tasks using very few examples - sometimes even just three to five. In contrast to conventional learning methods, which require thousands of training data, few-shot learning saves time and resources.
Imagine you want to teach a piece of software to recognise different types of exotic flowers. With traditional methods, you would have to show the AI thousands of photos of each flower. With few-shot learning, just a few images are enough for the software to recognise the flower independently in future.
This has great advantages: Less data means less effort in collecting and preparing training material. Few-shot learning is particularly useful wherever little data is available - for example, for rare diseases in medicine or for recognising new products in digital commerce.
The bottom line is that Few-Shot Learning helps companies to integrate AI applications into their processes faster and more efficiently and to react more flexibly to new developments.
Deutsch 
English (UK)