Scientists have developed a new machine-learning algorithm to help robots display appropriate social behaviour in interactions with humans.
Researchers at Brown University and Tufts University , US, have created a cognitive computational model of human norms in a representation that can be coded into machines.
They developed an algorithm that allows machines to learn norms in unfamiliar situations by drawing on human data.
The project, funded by the US Defence Advanced Research Projects Agency (DARPA), represents important progress towards the development of AI systems that can "intuit" how to behave in certain situations much the way people do. The new research could bring closer machines that emulate the best of human behaviour, researchers said.