Hands-On Mathematics for Deep Learning

Mean squared error

The mean squared error (MSE) is a measure of how good an estimator is and is a better indicator of this than the bias. We write it as follows:

However, sometimes, we use the root MSE, which is the square root of the MSE.

We can also express the MSE in terms of bias and variance, as follows:

Sometimes, when we are trying to get a low MSE, it is in our best interest to have a biased estimator with low variance. We call this the bias-variance trade-off.