Войти в систему

Home
    - Создать дневник
    - Написать в дневник
       - Подробный режим

LJ.Rossia.org
    - Новости сайта
    - Общие настройки
    - Sitemap
    - Оплата
    - ljr-fif

Редактировать...
    - Настройки
    - Список друзей
    - Дневник
    - Картинки
    - Пароль
    - Вид дневника

Сообщества

Настроить S2

Помощь
    - Забыли пароль?
    - FAQ
    - Тех. поддержка



Пишет Misha Verbitsky ([info]tiphareth)
@ 2014-08-18 12:32:00


Previous Entry  Add to memories!  Tell a Friend!  Next Entry
Настроение: sick
Музыка:Derived category of coherent sheaves and counting invariants
Entry tags:math

степень формы рациональна
Доказал такую простую штуку: если какая-то степень
симметрической или симплектической формы рациональна,
то она рациональна. Положу сюда, чтобы не потерялось.

\subsection{Rational symmetric and skew-symmetric forms}

For the purpose of this paper,
{\bf a rational tensor} on $\R^n$ is a tensor $\Theta$ such that
$c\Theta$ is rational, for some $c\in \R$.
Further on, we shall need the following two
propositions.

\hfill

\proposition
Let $V=\R^n$ be a vector space, and $g\in \Sym^2 V^*$ a bilinear symmetric
form on $V$ such that $g^k\in \Sym^{2k}V^*$ is rational. Then
$g$ is rational.

\hfill

{\bf Proof:} Let $x\in V$ be a vector such that $g(x,x)\neq 0$.
Consider the map $R_g:\; V \arrow \R$ mapping $y$ to $g^k(x,x,x, ..., x, y)$.
Clearly, $x^\bot=\ker R_g$. Therefore, for each rational $x$,
its orthogonal complement $x^\bot$ is also rational.
Gramm-Schmidt orthogonalization method allows one to
construct a rational orthogonal basis $x_i$ for $g$.
Modifying $g$ by a constant multiplier, we may assume that
$g(x_1, x_1)=\pm 1$. Then $g^k(x_i, x_i, x_1, ..., x_1)=g(x_i, x_i)$,
and this number must be rational, because $g^k(x_1, ..., x_1)$
is rational. \endproof

\hfill

\proposition
Let $V=\R^{2n}$ be a vector space, and $g\in \Lambda^2 V^*$
a bilinear, non-degenerate skew-symmetric
form on $V$ such that $\omega^k\in \Lambda^{2k}V^*$ is rational,
for some $k

[Error: Irreparable invalid markup ('<n$.>') in entry. Owner must fix manually. Raw contents below.]

Доказал такую простую штуку: если какая-то степень
симметрической или симплектической формы рациональна,
то она рациональна. Положу сюда, чтобы не потерялось.

<lj-cut>\subsection{Rational symmetric and skew-symmetric forms}

For the purpose of this paper,
{\bf a rational tensor} on $\R^n$ is a tensor $\Theta$ such that
$c\Theta$ is rational, for some $c\in \R$.
Further on, we shall need the following two
propositions.

\hfill

\proposition
Let $V=\R^n$ be a vector space, and $g\in \Sym^2 V^*$ a bilinear symmetric
form on $V$ such that $g^k\in \Sym^{2k}V^*$ is rational. Then
$g$ is rational.

\hfill

{\bf Proof:} Let $x\in V$ be a vector such that $g(x,x)\neq 0$.
Consider the map $R_g:\; V \arrow \R$ mapping $y$ to $g^k(x,x,x, ..., x, y)$.
Clearly, $x^\bot=\ker R_g$. Therefore, for each rational $x$,
its orthogonal complement $x^\bot$ is also rational.
Gramm-Schmidt orthogonalization method allows one to
construct a rational orthogonal basis $x_i$ for $g$.
Modifying $g$ by a constant multiplier, we may assume that
$g(x_1, x_1)=\pm 1$. Then $g^k(x_i, x_i, x_1, ..., x_1)=g(x_i, x_i)$,
and this number must be rational, because $g^k(x_1, ..., x_1)$
is rational. \endproof

\hfill

\proposition
Let $V=\R^{2n}$ be a vector space, and $g\in \Lambda^2 V^*$
a bilinear, non-degenerate skew-symmetric
form on $V$ such that $\omega^k\in \Lambda^{2k}V^*$ is rational,
for some $k<n$. Then $\omega$ is rational.

\hfill

{\bf Proof. Step 1:} Assume that $k= n-1$. Then
the map $v, v' \arrow \frac{\omega^k\wedge v \wedge v'}{\omega^n}$
defines a non-degenerate form on $V^*$, dual to $\omega$.
Since this form is rational, $\omega$ is also rational.

{\bf Step 2:} Now assume that $1< k< n-1$. Fix two rational
vectors $x, x'$ such that $\omega(x, x')\neq 0$. Rescaling
$\omega$, we may assume that $\omega(x, x')=1$. Then
for any $k+1$-dimensional subspace $W\subset V$ containing
$x, x'$ and such that $\omega\restrict W$ is non-degenerate,
$\omega\retsrict W$ is rational and takes rational values
on all rational vectors. Since any two vectors
$y, y'\in V$ are contained in appropriate $W$, this
implies that $\omega(y, y')$ is rational.
\endproof
</lj-cut>


(Добавить комментарий)


[info]don_oles
2014-08-18 13:46 (ссылка)
ахуеть.

(Ответить)


[info]wieiner_
2014-08-18 16:41 (ссылка)
интересна, а это можно ТЕХом прочитать?

(Ответить) (Ветвь дискуссии)


[info]hasek
2014-08-18 19:21 (ссылка)
В TeX-овской разметке же и написано.

(Ответить) (Уровень выше)