Understanding empirical probability densities in terms of Dirac measure

39 Views Asked by At

We are given a dataset in 3 dimension $$ X = \begin{bmatrix} 2 & 4 & 3\\ 1 & 0.5 & 0.8 \end{bmatrix} $$ and corresponding label vector is $Y = \begin{bmatrix} 0 \\ 1 \end{bmatrix}$. If one defines the joint empirical probability as $P(X, Y) = \frac{1}{n} = \sum_{i=1}^n \delta_{x_i, y_i}$ where n is number of samples (2) and $\delta$ being Dirac measure.

My question is, how exactly this equation expands in this case ?. What are the values of $\delta(x_i, y_i)$ above ?.