Microsoft and MIT have collaborated on a new model that can detect and help fill the gaps in AV technology. Whenever an automated vehicle (AV) doesn’t know how to behave in a situation, the AI part of the vehicle’s technology would compare what the vehicle was about to do with what a human driver would do.
A novel model developed by MIT and Microsoft researchers identifies instances in which autonomous systems have “learned” from training examples that don’t match what’s actually happening in the real world. Engineers could use this model to improve the safety of artificial intelligence systems, such as driverless vehicles and autonomous robots. […]