One of the main objectives of science and engineering is to predict the results of different decisions. In many cases, it is important to distinguish between aleatoric uncertainty, when different future situations are possible, and epistemic uncertainty, when only one situation is possible - but we do not know which one will occur. For example, in weather prediction, we largely deal with aleatoric uncertainty, because many details of the future weather pattern contain random components that cannot be predicted. On the other hand, when we are predicting the trajectory of an asteroid, we know that its trajectory is uniquely determined by its initial position and initial velocity, but since we do not know the exact values of this position and velocity, we cannot uniquely predict where the asteroid will be in a few months.
展开▼
