Non-symmetric polarization
Andreas Defant, Sunke Schl\"uters

TL;DR
This paper investigates the relationship between the norms of non-symmetric multilinear forms derived from homogeneous polynomials and the original polynomials, establishing bounds that depend logarithmically on the number of variables.
Contribution
It provides new estimates for the norms of non-symmetric multilinear forms associated with homogeneous polynomials, extending understanding of their behavior under various norms.
Findings
Norms of non-symmetric multilinear forms can be estimated by polynomial norms with a logarithmic factor.
For -norms with p<2, the logarithmic factor is unnecessary.
The results apply to a broad class of norms on complex vector spaces.
Abstract
Let be an -homogeneous polynomial in -complex variables . Clearly, has a unique representation in the form \begin{equation*} P(x)= \sum_{1 \leq j_1 \leq \dotsc \leq j_m \leq n} c_{(j_1, \dotsc, j_m)} \, x_{j_1} \dotsb x_{j_m} \,, \end{equation*} and the "~form \begin{equation*} L_P(x^{(1)}, \dotsc, x^{(m)})= \sum_{1 \leq j_1 \leq \dotsc \leq j_m \leq n} c_{(j_1, \dotsc, j_m)} \, x^{(1)}_{j_1} \dotsb x^{(m)}_{j_m} \end{equation*} satisfies for every . We show that, although in general is non-symmetric, for a large class of reasonable norms on the norm of on up to a logarithmic term can be estimated by the norm of on ; here denotes a universal…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
