Skip to main content
Log in

On the Adjoint of Linear Relations in Hilbert Spaces

  • Published:
Mediterranean Journal of Mathematics Aims and scope Submit manuscript

Abstract

Assume that \({\mathfrak {H}}\) and \({\mathfrak {K}}\) are two real or complex Hilbert spaces, A a linear relation from \({\mathfrak {H}}\) to \({\mathfrak {K}}\), and B a linear relation from \({\mathfrak {K}}\) to \({\mathfrak {H}}\), respectively. Necessary and sufficient conditions for B to be equal to the adjoint of A are provided. Several consequences are also presented. More precisely, new characterizations for closed, skew–adjoint, selfadjoint, normal linear relations, and generalized orthogonal projections are obtained.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Ammar, A., Jeribi, A., Saadaoui, B.: Frobenius-Schur factorization for multivalued \(2 \times 2\) matrices linear operator. Mediterr. J. Math. 14(1), 29 (2017). (Art. 29)

    Article  MathSciNet  Google Scholar 

  2. Arens, R.: Operational calculus of linear relations. Pac. J. Math. 11, 9–23 (1961)

    Article  MathSciNet  Google Scholar 

  3. Coddington, E.A.: Extension theory of formally normal and symmetric subspaces. Memoirs of the American Mathematical Society, no. 134 (1973)

  4. Coddington, E.A., de Snoo, H.S.V.: Positive selfadjoint extensions of positive symmetric subspaces. Math. Z. 159, 203–214 (1978)

    Article  MathSciNet  Google Scholar 

  5. Hassi, S., Sandovici, A., de Snoo, H.S.V., Winkler, H.: Form sums of nonnegative selfadjoint operators. Acta Math. Hung. 111, 81–105 (2006)

    Article  MathSciNet  Google Scholar 

  6. Hassi, S., Sandovici, A., de Snoo, H.S.V., Winkler, H.: A general factorization approach to the extension theory of nonnegative operators and relations. J. Oper. Theory 58, 351–386 (2007)

    MATH  Google Scholar 

  7. Hassi, S., Sandovici, A., de Snoo, H.S.V., Winkler, H.: Extremal extensions for the sum of nonnegative selfadjoint relations. Proc. Am. Math. Soc. 135, 3193–3204 (2007)

    Article  MathSciNet  Google Scholar 

  8. Hassi, S., de Snoo, H.S.V.: Factorization, majorization, and domination for linear relations. Ann. Univ. Sci. Bp. 58, 55–72 (2015)

    MathSciNet  MATH  Google Scholar 

  9. Hassi, S., de Snoo, H.S.V., Szafraniec, F.H.: Componentwise and canonical decompositions of linear relations, Dissertationes Mathematicae, vol. 465, pp. 59 (2009)

  10. Iftime, O.V., Roman, M., Sandovici, A.: A kernel representation of Dirac structures for infinite-dimensional systems. Math. Model. Nat. Phenom. 9(5), 295–308 (2014). https://doi.org/10.1051/mmnp/20149520

    Article  MathSciNet  MATH  Google Scholar 

  11. von Neumann, J.: Allgemeine Eigenwerttheorie Hermitescher Funktionaloperatoren. Math. Ann. 102, 49–131 (1930)

    Article  MathSciNet  Google Scholar 

  12. von Neumann, J.: Uber adjungierte Funktionaloperatoren. Ann. Math. (2) 33, 294–310 (1932)

    Article  MathSciNet  Google Scholar 

  13. Popovici, D., Sebestyén, Z.: On operators which are adjoint to each other. Acta Sci. Math. (Szeged) 80, 175–194 (2014)

    Article  MathSciNet  Google Scholar 

  14. Popovici, D., Sebestyén, Z., Tarcsay, Z.: On the sum between a closable operator T and a T-bounded operator. Ann. Univ. Sci. Bp. Sect. Math. 58, 95–104 (2015)

    MathSciNet  MATH  Google Scholar 

  15. Roman, M., Sandovici, A.: Generalized Timoshenko beam model for buckling and vibration of nanowires via Dirac structures on infinite dimensional Hilbert spaces (in preparation)

  16. Roman, M., Sandovici, A.: A factorization approach to the extension theory of the tensor product of nonnegative linear relations. Results Math. 72, 875–891 (2017)

    Article  MathSciNet  Google Scholar 

  17. Roman, M., Sandovici, A.: B-spectral theory of linear relations in complex Banach spaces. Publ. Math. Debr. 91(3–4), 455–466 (2017)

    Article  MathSciNet  Google Scholar 

  18. Roman, M., Sandovici, A.: A note on a paper by Nieminen. Results Math. 74, 73 (2019). https://doi.org/10.1007/s00025-019-1002-2

    Article  MathSciNet  MATH  Google Scholar 

  19. Roman, M., Sandovici, A.: The square root of nonnegative selfadjoint linear relations in Hilbert spaces. J. Oper. Theory 82(2), 357–367 (2019)

    MathSciNet  MATH  Google Scholar 

  20. Sandovici, A.: Self-adjointness and skew-adjointness criteria involving powers of linear relations. J. Math. Anal. Appl. 470(1), 186–200 (2019)

    Article  MathSciNet  Google Scholar 

  21. Sandovici, A.: Von Neumann’s theorem for linear relations. Linear Multilinear Algebra 66(9), 1750–1756 (2018)

    Article  MathSciNet  Google Scholar 

  22. Sandovici, A., Vasilescu, F.H.: Normal extensions of subnormal linear relations via quaternionic Cayley transforms. Monatshefte Math. 170(3–4), 437–463 (2013)

    Article  MathSciNet  Google Scholar 

  23. Sebestyén, Z.: On ranges of adjoint operators in Hilbert space. Acta Sci. Math. (Szeged) 46, 295–298 (1983)

    MathSciNet  MATH  Google Scholar 

  24. Sebestyén, Z., Stochel, J.: Restrictions of positive selfadjoint operators. Acta Sci. Math. (Szeged) 55, 149–154 (1991)

    MathSciNet  MATH  Google Scholar 

  25. Sebestyén, Z., Tarcsay, Z.: \(T^{*}T\) always has a positive selfadjoint extension. Acta Math. Hung. 135, 116–129 (2012)

    Article  MathSciNet  Google Scholar 

  26. Sebestyén, Z., Tarcsay, Z.: Characterizations of selfadjoint operators. Stud. Sci. Math. Hung. 50, 423–435 (2013)

    MathSciNet  MATH  Google Scholar 

  27. Sebestyén, Z., Tarcsay, Z.: Adjoint of sums and products of operators in Hilbert spaces. Acta Sci. Math. (Szeged) 82(1–2), 175–191 (2016)

    Article  MathSciNet  Google Scholar 

  28. Sebestyén, Z., Tarcsay, Z.: Operators having selfadjoint squares. Ann. Univ. Sci. Bp. Sect. Math. 58, 105–110 (2015)

    MathSciNet  MATH  Google Scholar 

  29. Sebestyén, Z., Tarcsay, Z.: A reversed Von Neumann theorem. Acta Sci. Math. (Szeged) 80, 659–664 (2014)

    Article  MathSciNet  Google Scholar 

  30. Sebestyén, Z., Tarcsay, Z.: Characterizations of essentially selfadjoint and skew-adjoint operators. Stud. Sci. Math. Hung. 52, 371–385 (2015)

    MATH  Google Scholar 

  31. Sebestyén, Z., Tarcsay, Z.: Adjoint of sums and products of operators in Hilbert spaces. Acta Sci. Math. (Szeged) 82, 175–191 (2016)

    Article  MathSciNet  Google Scholar 

  32. Sebestyén, Z., Tarcsay, Z.: On the adjoint of Hilbert space operators. Linear Multilinear Algebra. https://doi.org/10.1080/03081087.2018.1430120

  33. Tarcsay, Z.: Operator extensions with closed range. Acta Math. Hung. 135, 325–341 (2012)

    Article  MathSciNet  Google Scholar 

Download references

Acknowledgements

I would like to thank the reviewer for his/her thoughtful comments and efforts towards improving the present paper by pointing out some mistakes in the proofs of Theorems 7.3 and 8.2 and also for bringing into attention the paper [1].

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Adrian Sandovici.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

Appendix

The next result is useful with respect to the multiplication of \(2 \times 2\) matrices of linear relations.

Lemma 9.1

Assume that \({\mathfrak {H}}\) and \({\mathfrak {K}}\) are two real ar complex Hilbert spaces and let \(A \in {\mathfrak {L}}({\mathfrak {H}},{\mathfrak {K}})\) and \(A \in {\mathfrak {L}}({\mathfrak {K}},{\mathfrak {H}})\) The following identity holds true:

$$\begin{aligned} \left( \begin{array}{cc} I &{}\quad -B \\ A &{}\quad I \\ \end{array} \right) \left( \begin{array}{cc} I &{}\quad B \\ -A &{}\quad I \\ \end{array} \right) = \left( \begin{array}{cc} I+BA &{}\quad \mathrm{dom\,}B \times \mathrm{mul\,}B \\ \mathrm{dom\,}A \times \mathrm{mul\,}A &{}\quad I + AB \\ \end{array} \right) . \end{aligned}$$
(9.12)

Proof

Assume that \( \left\{ \left( \begin{array}{c} x \\ y \end{array} \right) , \left( \begin{array}{c} z \\ t \end{array} \right) \right\} \in \left( \begin{array}{cc} I &{}\quad -B \\ A &{}\quad I \\ \end{array} \right) \left( \begin{array}{cc} I &{}\quad B \\ -A &{}\quad I \\ \end{array} \right) \), so that

$$\begin{aligned} \left\{ \left( \begin{array}{c} x \\ y \end{array} \right) , \left( \begin{array}{c} a \\ b \end{array} \right) \right\} \in \left( \begin{array}{cc} I &{}\quad B \\ -A &{}\quad I \\ \end{array} \right) , \quad \left\{ \left( \begin{array}{c} a \\ b \end{array} \right) , \left( \begin{array}{c} z \\ t \end{array} \right) \right\} \in \left( \begin{array}{cc} I \,\,&{} -B \\ A \,\,&{} I \\ \end{array} \right) , \end{aligned}$$

for some \(a \in {\mathfrak {H}}\) and \(b \in {\mathfrak {K}}\). Therefore, \(a = x + y^{\prime }\) and \(b = -x^{\prime } +y\) for some \(\{x,x^{\prime }\} \in A\) and \(\{y,y^{\prime }\} \in B\). Also \(z = a-b^{\prime }\) and \(t = a^{\prime } + b\) for some \(\{a,a^{\prime }\} \in A\) and \(\{b,b^{\prime }\} \in B\) . Consequently, \(z = x + y^{\prime } - b^{\prime }\) and \(t = a^{\prime } - x^{\prime } + y\). Furthermore, \(\{x,x^{\prime }\} \in A\) implies that \(\{x,y-b\} \in A\) and:

$$\begin{aligned} \{y-b, y^{\prime }-b^{\prime }\} = \{y,y^{\prime }\} - \{b,b^{\prime }\} \in B \end{aligned}$$

so that \(\{x,y^{\prime }-b^{\prime }\} \in AB\), which further leads to:

$$\begin{aligned} \{x,x+y^{\prime }-b^{\prime }\} = \{x,z\} \in I + AB. \end{aligned}$$
(9.13)

Similar arguments shows that:

$$\begin{aligned} \{y,t\} \in I + BA. \end{aligned}$$
(9.14)

Since also:

$$\begin{aligned} \{y,0\} \in \mathrm{dom\,}B \times \mathrm{mul\,}B, \quad \{x,0\} \in \mathrm{dom\,}A \times \mathrm{mul\,}A, \end{aligned}$$
(9.15)

it follows from (9.13), (9.14) and (9.15) that:

$$\begin{aligned} \left\{ \left( \begin{array}{c} x \\ y \end{array} \right) , \left( \begin{array}{c} z \\ t \end{array} \right) \right\} \in \left( \begin{array}{cc} I + BA &{}\quad \mathrm{dom\,}B \times \mathrm{mul\,}B \\ \mathrm{dom\,}A \times \mathrm{mul\,}A &{}\quad I + AB \\ \end{array} \right) . \end{aligned}$$

Thus:

$$\begin{aligned} \left( \begin{array}{cc} I &{} -B \\ A &{} I \\ \end{array} \right) \left( \begin{array}{cc} I &{} B \\ -A &{} I \\ \end{array} \right) \subset \left( \begin{array}{cc} I+BA &{}\quad \mathrm{dom\,}B \times \mathrm{mul\,}B \\ \mathrm{dom\,}A \times \mathrm{mul\,}A &{}\quad I + AB \\ \end{array} \right) . \end{aligned}$$
(9.16)

Conversely, it will be shown that:

$$\begin{aligned} \left( \begin{array}{cc} I+BA &{} \mathrm{dom\,}B \times \mathrm{mul\,}B \\ \mathrm{dom\,}A \times \mathrm{mul\,}A &{} I + AB \\ \end{array} \right) \subset \left( \begin{array}{cc} I &{}\quad -B \\ A &{}\quad I \\ \end{array} \right) \left( \begin{array}{cc} I &{}\quad B \\ -A &{}\quad I \\ \end{array} \right) . \end{aligned}$$
(9.17)

Assume that \( \left\{ \left( \begin{array}{c} x \\ y \end{array} \right) , \left( \begin{array}{c} z \\ t \end{array} \right) \right\} \in \left( \begin{array}{cc} I + BA &{} \mathrm{dom\,}B \times \mathrm{mul\,}B \\ \mathrm{dom\,}A \times \mathrm{mul\,}A &{} I + AB \\ \end{array} \right) , \) so that \(z = x^{\prime } + m_{2}\) and \(t = y^{\prime } + m_{1}\) for some \(\{x,x^{\prime }\} \in I + BA\), \(\{y,y^{\prime }\} \in I + AB\), \(m_{1} \in \mathrm{mul\,}A\) and \(m_{2} \in \mathrm{mul\,}B\). It follows from \(\{x,x^{\prime }\} \in I + BA\) that \(\{x,x^{\prime }-x\} \in BA\), so that \(\{x,a\} \in A\), \(\{a,x^{\prime }-x\} \in B\) for some \(a \in {\mathfrak {K}}\). Also, it follows from \(\{y,y^{\prime }\} \in I+AB\) that \(\{y,y^{\prime }-y\} \in AB\), so that \(\{y,b\} \in B\), \(\{b,y^{\prime }-y\} \in A\) for some \(b \in {\mathfrak {H}}\). Then, clearly:

$$\begin{aligned} \left\{ \left( \begin{array}{c} x \\ y \end{array} \right) , \left( \begin{array}{c} x + b \\ -a + y \end{array} \right) \right\} \in \left( \begin{array}{cc} I &{}\quad B \\ -A &{}\quad I \\ \end{array} \right) . \end{aligned}$$
(9.18)

Furthermore, the following relations:

$$\begin{aligned} \{x+b,a+y^{\prime }-y+m_{1}\}= & {} \{x,a\} + \{b,y^{\prime }-y\} + \{0,m_{1}\} \in A, \end{aligned}$$
(9.19)
$$\begin{aligned} \{-a+y,b-x^{\prime }+x-m_{2}\}= & {} - \{a,x^{\prime }-x\} + \{y,b\} - \{0,m_{2}\} \in B,\nonumber \\ \end{aligned}$$
(9.20)

imply that:

$$\begin{aligned}&\left\{ \left( \begin{array}{c} x + b \\ -a + y \end{array} \right) , \left( \begin{array}{c} x+b-b+x^{\prime }-x+m_{2} \\ a+y^{\prime }-y+m_{1}-a+y \end{array} \right) \right\} \nonumber \\&\quad = \left\{ \left( \begin{array}{c} x + b \\ -a + y \end{array} \right) , \left( \begin{array}{c} z \\ t \end{array} \right) \right\} \in \left( \begin{array}{cc} I &{}\quad -B \\ A &{}\quad I \\ \end{array} \right) . \end{aligned}$$
(9.21)

A combination of (9.18) and (9.21) shows that:

$$\begin{aligned} \left\{ \left( \begin{array}{c} x \\ y \end{array} \right) , \left( \begin{array}{c} z \\ t \end{array} \right) \right\} \in \left( \begin{array}{cc} I &{}\quad -B \\ A &{}\quad I \\ \end{array} \right) \left( \begin{array}{cc} I &{}\quad B \\ -A &{}\quad I \\ \end{array} \right) , \end{aligned}$$

so that (9.17) has been proved. Finally, it follows from (9.16) and (9.17) that (9.12) holds true. \(\square \)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Sandovici, A. On the Adjoint of Linear Relations in Hilbert Spaces. Mediterr. J. Math. 17, 68 (2020). https://doi.org/10.1007/s00009-020-1503-y

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s00009-020-1503-y

Keywords

Mathematics Subject Classification

Navigation