I saw recently the following claim. Given any 19 points in $\mathbb{R}^3$ it is always possible to find a polynomial $p(x,y,z)$ with $\deg p \leq 3$ such that it vanishes in the previous points.
My question is why is this true (for the previous particular case) and where can I find a general version of the statement (given any $m$ points in $\mathbb{R}^n$ which is the least positive integer $d$ such that I can always find a polynomial $p$ that vanishes at the points with degree $\leq d$?).
Thank you in advance!
I am not sure, but this is just a guess.
Given $n$ points in $\Bbb R$, you can always find a polynomial (in one variable) of degree $n$ that vanishes at the given points. This follows from the Fundamental Theorem of Algebra.
With a polynomial of degree $n$, there are $n+1$ coefficients, and the "solution polynomial" is only defined up to a constant multiple.
The general form of a three variable polynomial is
$$a_1x^3+a_2y^3+a^3z^3+a_4x^2y+a_5x^2z+a_6y^2x+a_7y^2z+a_8z^2x+a_9z^2y+a_{10}xyz+a_{11}x^2+a_{12}y^2+a_{13}z^2+a_{14}xy+a_{15}yz+a_{16}zx+a_{17}x+a_{18}y+a_{19}z+a_{20}$$
So I would guess that, since there are $20$ coefficients, you can always make it vanish at any $19$ points.