Hilbert algebras are examined in the mathematical sub-area of functional analysis. They are algebras with an additional pre-Hilbert space structure, which explains the name Hilbert algebra . On the completion can be Von Neumann algebras construct, ultimately leading to a characterization of the semi-finite von Neumann algebras leads. A generalization of this construction, which applies to every Von-Neumann algebra, leads to the concept of the generalized Hilbert algebra and is the starting point of the Tomita-Takesaki theory .
Definitions
A Hilbert algebra is an associative algebra over the field of complex numbers together with an involution and a scalar product , which makes it a pre-Hilbert space, so that the following conditions are met:
![\ langle \ cdot | \ cdot \ rangle](https://wikimedia.org/api/rest_v1/media/math/render/svg/d36ddc87338be38721a2d721d1a6cc9ab294474a)
![A.](https://wikimedia.org/api/rest_v1/media/math/render/svg/7daff47fa58cdfd29dc333def748ff5fa4c923e3)
-
for all
-
for all
- For each , the mapping is continuous in the norm topology defined by the scalar product .
![x \ in U](https://wikimedia.org/api/rest_v1/media/math/render/svg/c32ddcb2941216f2980b950ce969dc15cba26906)
![A \ rightarrow A, \, y \ mapsto xy](https://wikimedia.org/api/rest_v1/media/math/render/svg/99351723cc44e7df7366b8bd4642d8f579dd88df)
- Out for everyone follows .
![\ langle xy | z \ rangle = 0](https://wikimedia.org/api/rest_v1/media/math/render/svg/69351a2fcc25f304923f20039e37f155257652c5)
![x, y \ in A](https://wikimedia.org/api/rest_v1/media/math/render/svg/9fd3c2bd17578ba23f55a299668cd7accc5c7a9c)
![z = 0](https://wikimedia.org/api/rest_v1/media/math/render/svg/b92bfc06485cc90286474b14a516a68d8bfdd7b3)
In the following is the Hilbert space , which results as the completion of . From the first condition it follows that the involution continues to a continuous, conjugate linear mapping for which
![H](https://wikimedia.org/api/rest_v1/media/math/render/svg/75a9edddcca2f782014371f75dca39d7e13a9c1b)
![A.](https://wikimedia.org/api/rest_v1/media/math/render/svg/7daff47fa58cdfd29dc333def748ff5fa4c923e3)
![J: H \ rightarrow H](https://wikimedia.org/api/rest_v1/media/math/render/svg/d2665ec77c06dc6ba0e7ce575efbc73682b13cff)
-
and for everyone![\ langle Jx | Jy \ rangle = \ langle y | x \ rangle](https://wikimedia.org/api/rest_v1/media/math/render/svg/218a1e091e6d2d6397a2086359054b8541ffc031)
holds, one calls the canonically defined involution on .
![J](https://wikimedia.org/api/rest_v1/media/math/render/svg/359e4f407b49910e02c27c2f52e87a36cd74c053)
![A.](https://wikimedia.org/api/rest_v1/media/math/render/svg/7daff47fa58cdfd29dc333def748ff5fa4c923e3)
![H](https://wikimedia.org/api/rest_v1/media/math/render/svg/75a9edddcca2f782014371f75dca39d7e13a9c1b)
The maps and continue for each to form continuous linear operators and such that:
![y \ mapsto xy](https://wikimedia.org/api/rest_v1/media/math/render/svg/392784a08e9de6ece37a41f0cb30c136f650a88c)
![y \ mapsto yx](https://wikimedia.org/api/rest_v1/media/math/render/svg/8253f4be69932c185184495f619559d31c857e63)
![x \ in A](https://wikimedia.org/api/rest_v1/media/math/render/svg/27bcc9b2afb295d4234bc294860cd0c63bcad2ca)
![U_ {x}: H \ rightarrow H](https://wikimedia.org/api/rest_v1/media/math/render/svg/c7d0e43312219c5e53b579ff55fd5bcce181af84)
![V_ {x}: H \ rightarrow H](https://wikimedia.org/api/rest_v1/media/math/render/svg/b80163a4ca3b7f9ed5523782d2268ceb5117414b)
-
is an involutive homomorphism
-
is an involutive antiomomorphism
-
for all
-
and for everyone![JV_ {x} J = U _ {{x ^ {*}}}](https://wikimedia.org/api/rest_v1/media/math/render/svg/01dbaeac61560d7cdeb47ed2ff2384b37f9680ca)
The completed cases concerning. The weak operator topologies of and be with and called and called the left-associated or right-associated von Neumann algebra . To prove that these are actually Von Neumann algebras, in particular that these algebras contain the identity , one needs the fourth condition of the above definition.
![\ {U_ {x} | x \ in A \}](https://wikimedia.org/api/rest_v1/media/math/render/svg/597e4e6b4ec65551de85705578218297d573c31a)
![\ {V_ {x} | x \ in A \}](https://wikimedia.org/api/rest_v1/media/math/render/svg/f12e0e97fec55e935dfd6cc62e8235b4c93a7a9a)
![U = U (A)](https://wikimedia.org/api/rest_v1/media/math/render/svg/9f2fbb3090212da4839a1319519e936842441f97)
![V = V (A)](https://wikimedia.org/api/rest_v1/media/math/render/svg/68a41ccbc8ccea994afae1072d661c178dcba115)
![A.](https://wikimedia.org/api/rest_v1/media/math/render/svg/7daff47fa58cdfd29dc333def748ff5fa4c923e3)
![{\ mathrm {id}} _ {H}](https://wikimedia.org/api/rest_v1/media/math/render/svg/d589a680231db11d1fe8b27a5285768dbe462ab8)
A von Neumann algebra is called a standard von Neumann algebra if it is of the form .
![U (A) =](https://wikimedia.org/api/rest_v1/media/math/render/svg/326ff423c43e9fe00a63d899c6fe3adff16c2bf6)
example
The H * -algebra algebra of the Hilbert-Schmidt operators on a Hilbert space is a Hilbert algebra. Denotes the conjugated Hilbert space , then is isomorphic to the Hilbert space tensor product . For is the one-dimensional operator if the scalar product is linear in the first component and conjugate linear in the second. Then
![A.](https://wikimedia.org/api/rest_v1/media/math/render/svg/7daff47fa58cdfd29dc333def748ff5fa4c923e3)
![H](https://wikimedia.org/api/rest_v1/media/math/render/svg/75a9edddcca2f782014371f75dca39d7e13a9c1b)
![\ overline {H}](https://wikimedia.org/api/rest_v1/media/math/render/svg/21e3c5eac166e464406970ed2cadc14fa7345da4)
![\ overline {H} \ otimes H](https://wikimedia.org/api/rest_v1/media/math/render/svg/2bca21faac3951aca9de90c2b313d160a6c47d5c)
![\ xi, \ eta \ in H](https://wikimedia.org/api/rest_v1/media/math/render/svg/d4072f5ffd013142edb895b5c2b38c30e565935b)
![\ xi \ otimes \ eta](https://wikimedia.org/api/rest_v1/media/math/render/svg/ba173831fccdc3af4acaa08087fd76aa1d815e7d)
![H \ rightarrow H, (\ xi \ otimes \ eta) \ zeta: = \ langle \ zeta, \ eta \ rangle \ xi](https://wikimedia.org/api/rest_v1/media/math/render/svg/02d59c53db1d14e14b64d2409f9a50401e146d3e)
![(\ xi _ {1} \ otimes \ eta _ {1}) (\ xi _ {2} \ otimes \ eta _ {2}) \ zeta = (\ xi _ {1} \ otimes \ eta _ {1} ) (\ langle \ zeta, \ xi _ {2} \ rangle \ eta _ {2}) = \ langle \ zeta, \ xi _ {2} \ rangle \ langle \ eta _ {2}, \ xi _ {1 } \ rangle \ eta _ {1} = \ langle \ eta _ {2}, \ xi _ {1} \ rangle (\ xi _ {2} \ otimes \ eta _ {1}) \ zeta](https://wikimedia.org/api/rest_v1/media/math/render/svg/8eb7eca3efea13b9c151861b4f1b1d8e77c75745)
and therefore
-
,
so
![U _ {{\ xi _ {1} \ otimes \ eta _ {1}}} (\ xi _ {2} \ otimes \ eta _ {2}) = \ langle \ eta _ {2}, \ xi _ {1 } \ rangle (\ xi _ {2} \ otimes \ eta _ {1}) = \ xi _ {2} \ otimes (\ langle \ xi _ {1}, \ eta _ {2} \ rangle \ eta _ { 1})](https://wikimedia.org/api/rest_v1/media/math/render/svg/fa9916e81b83a8ee85c994ccf4a377a3edee1e24)
-
,
where the tensor product denoted by the slash is the tensor product for Von Neumann algebras . One reads from it
![{\ displaystyle U (A) = \ mathbb {C} \ cdot 1 _ {\ overline {H}} \, {\ overline {\ otimes}} \, L (H)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/5546341a451e89a8a7d307df9c4a85d47a9280f6)
because the linear combinations from the one-dimensional operators lie close together in the respective algebras.
![\ xi _ {1} \ otimes \ eta _ {1}](https://wikimedia.org/api/rest_v1/media/math/render/svg/a9c75dba4862f2551f324aab24fbd61bcc3b3068)
Semi-infinite Von Neumann algebras
The Von Neumann algebras, which appear as the left-associated Von Neumann algebras of Hilbert algebras, are precisely the semi-finite Von Neumann algebras . If a Hilbert algebra is given, then there is a semi-finite, normal, faithful trace that makes a semi-finite Von Neumann algebra. Conversely, if a semi-finite Von Neumann algebra with such a trace is a Hilbert algebra with the scalar product defined by , whose left-associated Von Neumann algebra is too isomorphic.
![A.](https://wikimedia.org/api/rest_v1/media/math/render/svg/7daff47fa58cdfd29dc333def748ff5fa4c923e3)
![\ phi (U_ {a} ^ {*} U_ {a}): = \ langle a | a \ rangle](https://wikimedia.org/api/rest_v1/media/math/render/svg/837b1e47e9a4b9ec57f153f729d9320ef5dd72cf)
![U (A)](https://wikimedia.org/api/rest_v1/media/math/render/svg/fee555e9e7d76b5ab3b7889816fcacafce0651ad)
![U](https://wikimedia.org/api/rest_v1/media/math/render/svg/458a728f53b9a0274f059cd695e067c430956025)
![A: = \ {a \ in U | \ phi (a ^ {*} a) <\ infty, \ phi (aa ^ {*}) <\ infty \}](https://wikimedia.org/api/rest_v1/media/math/render/svg/cce5bf36ebf8e316be1b3282abf4de4f592616b6)
![\ langle a | b \ rangle: = \ phi (b ^ {*} a)](https://wikimedia.org/api/rest_v1/media/math/render/svg/ed2a21aea96f1764c9e74ea19cc943e77dee8a35)
![U](https://wikimedia.org/api/rest_v1/media/math/render/svg/458a728f53b9a0274f059cd695e067c430956025)
Generalized Hilbert Algebra
A generalized Hilbert algebra is an associative algebra over the field of complex numbers together with an involution and a scalar product , which makes it a pre-Hilbert space, so that the following conditions are met:
![A.](https://wikimedia.org/api/rest_v1/media/math/render/svg/7daff47fa58cdfd29dc333def748ff5fa4c923e3)
![{\ displaystyle \ mathbb {C}}](https://wikimedia.org/api/rest_v1/media/math/render/svg/f9add4085095b9b6d28d045fd9c92c2c09f549a7)
![x \ mapsto x ^ {*}](https://wikimedia.org/api/rest_v1/media/math/render/svg/fdcdf3ac500006405ee4d39c52f9c7617f82c64b)
![\ langle \ cdot | \ cdot \ rangle](https://wikimedia.org/api/rest_v1/media/math/render/svg/d36ddc87338be38721a2d721d1a6cc9ab294474a)
![A.](https://wikimedia.org/api/rest_v1/media/math/render/svg/7daff47fa58cdfd29dc333def748ff5fa4c923e3)
- The mapping is a closable , conjugate-linear operator in the completion .
![x \ mapsto x ^ {*}](https://wikimedia.org/api/rest_v1/media/math/render/svg/fdcdf3ac500006405ee4d39c52f9c7617f82c64b)
![H](https://wikimedia.org/api/rest_v1/media/math/render/svg/75a9edddcca2f782014371f75dca39d7e13a9c1b)
-
for all
- For each , the mapping is continuous in the norm topology defined by the scalar product .
![x \ in U](https://wikimedia.org/api/rest_v1/media/math/render/svg/c32ddcb2941216f2980b950ce969dc15cba26906)
![A \ rightarrow A, \, y \ mapsto xy](https://wikimedia.org/api/rest_v1/media/math/render/svg/99351723cc44e7df7366b8bd4642d8f579dd88df)
- Out for everyone follows .
![\ langle xy | z \ rangle = 0](https://wikimedia.org/api/rest_v1/media/math/render/svg/69351a2fcc25f304923f20039e37f155257652c5)
![x, y \ in A](https://wikimedia.org/api/rest_v1/media/math/render/svg/9fd3c2bd17578ba23f55a299668cd7accc5c7a9c)
![z = 0](https://wikimedia.org/api/rest_v1/media/math/render/svg/b92bfc06485cc90286474b14a516a68d8bfdd7b3)
Generalized Hilbert algebras are also called left Hilbert algebras.
Hilbert algebras are generalized Hilbert algebras. For this purpose it is necessary to show that the picture , is lockable, ie from and already follows. For each one follows a Hilbert algebra using the first defining property
![A \ rightarrow A](https://wikimedia.org/api/rest_v1/media/math/render/svg/435207fa671140cb134a72c3d2185ef44620a56d)
![x \ mapsto x ^ {*}](https://wikimedia.org/api/rest_v1/media/math/render/svg/fdcdf3ac500006405ee4d39c52f9c7617f82c64b)
![x_ {n} \ rightarrow 0](https://wikimedia.org/api/rest_v1/media/math/render/svg/91a1d7df5a4eb72f4dc6f8828b6ac06e15443a7e)
![x_ {n} ^ {*} \ rightarrow z](https://wikimedia.org/api/rest_v1/media/math/render/svg/a4ccdc183298c87c9707276e7b9bb7a2c026e287)
![z = 0](https://wikimedia.org/api/rest_v1/media/math/render/svg/b92bfc06485cc90286474b14a516a68d8bfdd7b3)
![y \ in A](https://wikimedia.org/api/rest_v1/media/math/render/svg/7967ddfa570a9b0cc13b4785f57407d2526cf292)
![\ langle z, y \ rangle = \ lim _ {{n \ to \ infty}} \ langle x_ {n} ^ {*}, y \ rangle = \ lim _ {{n \ to \ infty}} \ langle y ^ {*}, x_ {n} \ rangle = \ langle y ^ {*}, 0 \ rangle = 0](https://wikimedia.org/api/rest_v1/media/math/render/svg/7ddded6060bd35d1b69085880ecc20739a9a94ad)
and therefore , because was arbitrary, that is, is perpendicular to a dense subset of the completion.
![z = 0](https://wikimedia.org/api/rest_v1/media/math/render/svg/b92bfc06485cc90286474b14a516a68d8bfdd7b3)
![y \ in A](https://wikimedia.org/api/rest_v1/media/math/render/svg/7967ddfa570a9b0cc13b4785f57407d2526cf292)
![z](https://wikimedia.org/api/rest_v1/media/math/render/svg/bf368e72c009decd9b6686ee84a375632e11de98)
As above, the displays put on operators to continue their weakly closed hull forms the left-associated von Neumann algebra . The end of the figure is called the Sharp operator, which is why many authors write the involution with the Sharp character #. Its polar decomposition leads to the formulas that are described in the article on Tomita-Takesaki theory .
![y \ mapsto xy](https://wikimedia.org/api/rest_v1/media/math/render/svg/392784a08e9de6ece37a41f0cb30c136f650a88c)
![U_ {x}](https://wikimedia.org/api/rest_v1/media/math/render/svg/dc13aedb28dbebb2c1f22b8f82d9d36a3af8a9d1)
![H](https://wikimedia.org/api/rest_v1/media/math/render/svg/75a9edddcca2f782014371f75dca39d7e13a9c1b)
![S.](https://wikimedia.org/api/rest_v1/media/math/render/svg/4611d85173cd3b508e67077d4a1252c9c05abca2)
![x \ mapsto x ^ {*}](https://wikimedia.org/api/rest_v1/media/math/render/svg/fdcdf3ac500006405ee4d39c52f9c7617f82c64b)
![\ textstyle S = J \ Delta](https://wikimedia.org/api/rest_v1/media/math/render/svg/7d1e18b64c5823dfe7e2b90945ae48d229ace46b)
Any Von Neumann algebra has a faithful, normal, semi-infinite weight . Then there is
a generalized Hilbert algebra whose left-associated Von Neumann algebra is too isomorphic.
![\omega](https://wikimedia.org/api/rest_v1/media/math/render/svg/48eff443f9de7a985bb94ca3bde20813ea737be8)
![A: = \ {a \ in U | \, \ omega (a ^ {*} a) <\ infty, \ omega (aa ^ {*}) <\ infty \}](https://wikimedia.org/api/rest_v1/media/math/render/svg/8e53834495428d230b0cd742cb42e867363dca7a)
![U](https://wikimedia.org/api/rest_v1/media/math/render/svg/458a728f53b9a0274f059cd695e067c430956025)
Individual evidence
-
^ Jacques Dixmier : Von Neumann algebras. North-Holland, Amsterdam 1981, ISBN 0-444-86308-7 , Chapter I.5.1: Definition of Hilbert algebras.
-
^ Jacques Dixmier: Von Neumann algebras. North-Holland, Amsterdam 1981, ISBN 0-444-86308-7 , Chapter I, §5, Paragraph 5, Definition 7.
-
^ Jacques Dixmier: Von Neumann algebras. North-Holland Publishing, Amsterdam et al. 1981 (North-Holland Mathematical Library, Volume 27), ISBN 0-444-86308-7 , Part I, Chap. 6, para. 5: Normal traces on L (H).
-
^ Jacques Dixmier: Von Neumann algebras. North-Holland Publishing, Amsterdam et al. 1981 (North-Holland Mathematical Library, Volume 27), ISBN 0-444-86308-7 , Part I, Chap. 6, para. 2, Theorem 1 and Theorem 2.
-
^ M. Takesaki : Tomita's theory of modular Hilbert-algebras and its applications. Lecture Notes in Mathematics, Volume 128, Springer-Verlag 1970, §2.
-
^ RV Kadison , JR Ringrose : Fundamentals of the Theory of Operator Algebras II. Academic Press, 1983, ISBN 0-12-393302-1 , definition 9.2.41.
-
^ RV Kadison , JR Ringrose : Fundamentals of the Theory of Operator Algebras II , Academic Press (1983), ISBN 0-12-393302-1 , sentence 9.2.40.