Interactive demo · Foundations of Deep Learning · Chapter 2

Gaussian maximum likelihood and the bias of the variance

For independent data the likelihood is the product of the densities at the data points, ∏n N(xn | μ, σ2). Its maximum is at the sample mean and the sample variance about it. That variance estimate is measured from a centre fitted to the same data, so it is too small on average by the factor (N − 1)/N: a small, exactly computable example of how fitting and evaluating on the same data flatters a model.

View

Mean μ

Standard deviation σ

Number of points N

Points per data set N

Number of data sets

Scoreboard

Try these

In the lecture notes