Perfectness conjecture for preorderings of matrix polynomials

Let cmathcalB=cmathbbR[x1,,xn]cmathcal{B}=cmathbb{R}[x_1,\cdots,x_n], let S=cbracep1,,pmcbracecsubsetcmathcalBS=cbrace p_1,\cdots,p_mcbracecsubsetcmathcal{B}, and let TST_S be the preordering generated by SS. Set cmathcalA=MN(cmathcalB)cmathcal{A}=M_N(cmathcal{B}) and let cmathfraktrcolonMN(cmathcalB)ctocmathcalBcmathfrak{tr}colon M_N(cmathcal{B})ctocmathcal{B} be the normalized trace.

Perfectness conjecture. The induced quadratic module satisfies

cmathopIndcmathfraktrTS=TSccdotcSigmaMN(cmathcalB)2,cmathop{\operatorname{Ind}}^{cmathfrak{tr}} T_S=T_SccdotcSigma M_N(cmathcal{B})^2,

i.e. TST_S is a perfect quadratic module.

If TST_S is saturated, this is equivalent to representing every matrix polynomial positive semidefinite on KSK_S as csumicalphaiAiAicsum_icalpha_i A_i^*A_i with AicinMN(cmathcalB)A_icin M_N(cmathcal{B}) and calphaicinTScalpha_icin T_S. The representation is known in several one-variable and a particular two-variable case, but the conjecture remains open in general.

Sources & referencesView supporting material

Primary source

Jaka Cimpric and Yurii Savchuk, “Induced quadratic modules in *-algebras”, arXiv:1201.1374 (2014).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.