Root Mean Squared Errorn.
ruːt miːn skwɛːd ˈɛrə
A type of distance metric which is a variation of the Mean Squared Error metric. It applies a square root to the MSE distance, mitigating some of the effects of large errors, giving:
RMSE is used primarily in statistics to compare data sets, for example when testing the accuracy of predictions against a collection of measurements. RMSE employs non-symmetric penalisation, in that any worsening in accuracy is given extra weight compared to improvements of identical magnitude. RMSE is much more sensitive to large errors than the Mean Absolute Error, as all errors are squared prior to computing the average.