AI news: Neural network learns when it should not be trusted – '99% won't cut it'

AI experts developed a method for modelling the machine's confidence level based on the quality of the available data.MIT engineers expect this advance may eventually save lives, as deep learning is now widely deployed in everyday ways.For example, a network's level of certainty can be the difference between an autonomous vehicle determining between a clear crossroad and "it's probably clear, so stop just in case."This It can be used to assess products that rely on learned models."By AI analyst adds how previous approaches to uncertainty analysis are based on Bayesian deep learning."We