recover original decision variables from SOS program

[Hank Yang]

I am new to YALMIP and I am following the online guide for SOS programming:

https://yalmip.github.io/tutorial/sumofsquaresprogramming/

In the last example for finding a lower bound for a function f(x) using SOS, I am wondering how can I recover the original variable 'x' that achieves the lower bound of f(x)? The sossolve function only returns the lower bound now.

[Johan Löfberg]

sum-of-squares decompositions don't extract solutions, they only compute parameters such that an infinite dimensional polynomial constraint holds. To try to extract solutions from semidefinite relaxations based polynomials, you would use the solvemoment module