Answers are grounded in this paper and cite their sources.
For [Formula: see text], the cone [Formula: see text] of positive semidefinite (PSD) [Formula: see text]-ary [Formula: see text]-ic forms (i.e., homogeneous polynomials with real coefficients in [Formula: see text] variables of degree [Formula: see text]) contain the cone [Formula: see text] of those that are representable as finite sums of squares (SOS) of [Formula: see text]-ary [Formula: see text]-ic forms. Hilbert’s 1888 theorem states that [Formula: see text] exactly in the Hilbert cases [Formula: see text] with [Formula: see text] or [Formula: see text] or (3,4). For the non-Hilbert cases, we examined in a preceeding paper a specific cone filtration [Formula: see text] along [Formula: see text] projective varieties containing the Veronese variety via the Gram matrix method. Here, [Formula: see text] was the dimension of the real vector space of [Formula: see text]-ary [Formula: see text]-ic forms. In particular, we computed the number [Formula: see text] of strictly separating intermediate cones (i.e., [Formula: see text] such that [Formula: see text]) for the cases (3,6) and [Formula: see text]. In this paper, first, we generalize these findings to any non-Hilbert case by identifying each strict inclusion in the above specific cone filtration. This allows us to give a refinement of Hilbert’s 1888 theorem by computing [Formula: see text] in any non-Hilbert case. The cone filtration thus reduces to a specific cone subfiltration [Formula: see text] in which each inclusion is strict. Second, we show that each [Formula: see text], and hence each strictly separating [Formula: see text], fails to be a spectrahedral shadow.
No takes yet. Share an insight, caveat, or question.
Goel et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: