All Questions
Tagged with sums-of-squares real-algebraic-geometry
7 questions
1
vote
0
answers
68
views
Lower bounds on lengths of sum-of-squares representations of particular polynomials
I am looking for literature on the problem of finding minimal (in the sense of number of terms) sum-of-squares representations of particular non-negative multivariate polynomials with rational ...
1
vote
1
answer
268
views
Can I prove that a polynomial representing the 4th moment of a weighted-sum of random variables is a sos?
I am looking at the 4th central moment of a weighted-sum of correlated random variables, which takes the form
$$\mu_4 = \sum_{i,j,k,l=1}^n w_i w_j w_k w_l \mu_{ijkl}$$
where $\mu_{ijkl}$ are the ...
11
votes
0
answers
363
views
Is there yet an example of a non-negative convex polynomial that cannot be written as a sum-of-squares?
I have read that it remains an open question, whether an example can be constructed of a non-negative convex polynomial that cannot be written as a sum-of-squares. My reading includes the following ...
3
votes
0
answers
53
views
Points of intersection of summand of sums of squares of real polynomials
$\newcommand\R{\mathbb R}
\newcommand\Q{\mathbb Q}
$I am thinking of something related to Blekhermans 2012 paper Nonnegative Polynomials and Sums of Squares (Journal of the AMS, 25, 2012, 617-635).
...
4
votes
1
answer
113
views
Specific quaternary quartic that is positive semi-definite but not sum of squares
Does there exist a quaternary quartic $f$ (a form in $\mathbb{R}[x_1,x_2,x_3,x_4]$ of degree $4$), which is positive semi-definite ($f \geq 0$ on $\mathbb{R}^4$) but not a sum of squares, such that ...
2
votes
0
answers
187
views
Sums of hermitian squares in free abelian group algebras and real positive semidefinite matrices
A little context for the following question, first. As Noah Stein notes in a comment below, the present question is closely related to the free semialgebraic geometry studied by Helton and his ...
3
votes
1
answer
159
views
Sum of Squares Length of a Product
Let $n \geq 2$. Let $g_1, \ldots , g_{n-1} \in \mathbb{R}[x_1,\ldots,x_n]$ such that $q=g_1^2+\ldots +g_{n-1}^2$ is not divisible by $p=x_1^2+\ldots +x_n^2$. Let $m \geq 1$ be the smallest integer ...