root mean square

Terms from Artificial Intelligence: humans at the heart of algorithms

The glossary is being gradually proof checked, but currently has many typos and misspellings.

The root mean square (RMS) is the square root of the arithmetic mean of the squares of sequence of values . That is, if vi is a sequence of values:

        RMS   =   &sqrt ( ( &Sigma vi2 ) / N )

The RMS (or the plain mean square error) is often used as a form of {{accuracy measure}. In this case the values are differences between true value (say xi) and prediction or estimate (say pi), so that the RMS error is:

        RMS   =   &sqrt ( ( &Sigma (xi − pi)2 ) / N )

Also known as RMS