\(\DeclarePairedDelimiterX{\Set}[2]{\{}{\}}{#1 \nonscript\;\delimsize\vert\nonscript\; #2}\) \( \DeclarePairedDelimiter{\set}{\{}{\}}\) \( \DeclarePairedDelimiter{\parens}{\left(}{\right)}\) \(\DeclarePairedDelimiterX{\innerproduct}[1]{\langle}{\rangle}{#1}\) \(\newcommand{\ip}[1]{\innerproduct{#1}}\) \(\newcommand{\bmat}[1]{\left[\hspace{2.0pt}\begin{matrix}#1\end{matrix}\hspace{2.0pt}\right]}\) \(\newcommand{\barray}[1]{\left[\hspace{2.0pt}\begin{matrix}#1\end{matrix}\hspace{2.0pt}\right]}\) \(\newcommand{\mat}[1]{\begin{matrix}#1\end{matrix}}\) \(\newcommand{\pmat}[1]{\begin{pmatrix}#1\end{pmatrix}}\) \(\newcommand{\mathword}[1]{\mathop{\textup{#1}}}\)
Needs:
Real Matrix-Matrix Products
Real Vectors
Matrix Trace
Multivariate Real Polynomials
Needed by:
Quadratic Form Inequalities
Real Positive Semidefinite Matrices
Links:
Sheet PDF
Graph PDF

Quadratic Forms

Definition

A quadratic form is a multivariate polynomial each term of which has degree two.

Representation

Let $f: \R ^n \to \R $ be a quadratic form. There exists a matrix $A \in \R ^{n \times n}$ so that

\[ x^\top A x = \sum_{i,j} A_{ij}x_{i}x_{j}. \]

Suppose $A$ is not symmetric. Then

\[ f(x) = f(x)/2 + f(x)/2 = \frac{1}{2}x^\top Ax + \frac{1}{2}x^\top A^\top x = x^\top (\frac{1}{2}(A + A^\top ))x. \]

Define $B = 1/2 (A^\top + A)$. We call $B$ the symmetric part of $A$. Since every matrix $A$ has a symmetric part, we can always assume that the matrix for a quadratic form is symmetric. If it is not, replace it with its symmetric part, obtaining the same function.

Under trace

Observe that $\tr x^\top A x = \tr A x x^\top = \tr xx^\top A$.

Copyright © 2023 The Bourbaki Authors — All rights reserved — Version 13a6779cc About Show the old page view