Trace class
In mathematics, specifically functional analysis, a trace-class operator is a linear operator for which a trace may be defined, such that the trace is a finite number independent of the choice of basis used to compute the trace.
≈ 14 min read
Names in other languages
No names in other languages are held for this entry yet.
Article bodies are English and are not machine-translated. A name in another language is not a translation of the article.
In mathematics, specifically functional analysis, a trace-class operator is a linear operator for which a trace may be defined, such that the trace is a finite number independent of the choice of basis used to compute the trace. This trace of trace-class operators generalizes the trace of matrices studied in linear algebra. All trace-class operators are compact operators.
In quantum mechanics, quantum states are described by density matrices, which are certain trace class operators.[1]
Trace-class operators are essentially the same as nuclear operators, though many authors reserve the term "trace-class operator" for the special case of nuclear operators on Hilbert spaces and use the term "nuclear operator" in more general topological vector spaces (such as Banach spaces).
Definition
edit edit sourceLet H be a separable Hilbert space, {ek}k=1∞ an orthonormal basis and A:H→H a positive bounded linear operator on H. The trace of A is denoted by Tr(A) and defined as[2][3]
- Tr(A)=∑k=1∞⟨Aek,ek⟩,
independent of the choice of orthonormal basis. A (not necessarily positive) bounded linear operator T:H→H is called trace class if and only if
- Tr(|T|)<∞,
where |T|:=T∗T denotes the positive-semidefinite Hermitian square root.[4]
The trace-norm of a trace class operator T is defined as ‖T‖1:=Tr(|T|). One can show that the trace-norm is a norm on the space of all trace class operators B1(H) and that B1(H), with the trace-norm, becomes a Banach space.
When H is finite-dimensional, every (positive) operator is trace class. For A this definition coincides with that of the trace of a matrix. If H is complex, then A is always self-adjoint (i.e. A=A∗=|A|) though the converse is not necessarily true.[5]
Equivalent formulations
edit edit sourceGiven a bounded linear operator T:H→H, each of the following statements is equivalent to T being in the trace class:
- Tr(|T|)=∑k⟨|T|ek,ek⟩ is finite for every orthonormal basis (ek)k of H.[2]
- T is a nuclear operator.[6][7]
- There exist two orthogonal sequences (xi)i=1∞ and (yi)i=1∞ in H and positive real numbers (λi)i=1∞ in ℓ1 such that ∑i=1∞λi<∞ and
- x↦T(x)=∑i=1∞λi⟨x,xi⟩yi,∀x∈H,
- where (λi)i=1∞ are the singular values of T (or, equivalently, the eigenvalues of |T|), with each value repeated as often as its multiplicity.[8]
- There exist two orthogonal sequences (xi)i=1∞ and (yi)i=1∞ in H and positive real numbers (λi)i=1∞ in ℓ1 such that ∑i=1∞λi<∞ and
- T is a compact operator with Tr(|T|)<∞.
- If T is trace class then[9]
- ‖T‖1=sup{|Tr(CT)|:‖C‖≤1 and C:H→H is a compact operator }.
- If T is trace class then[9]
- T is an integral operator.[10]
- T is equal to the composition of two Hilbert-Schmidt operators.[11]
- |T| is a Hilbert-Schmidt operator.[11]
Examples
edit edit sourceSpectral theorem
edit edit sourceLet T be a bounded self-adjoint operator on a Hilbert space. Then T2 is trace class if and only if T has a pure point spectrum with eigenvalues {λi(T)}i=1∞ such that[12]
- Tr(T2)=∑i=1∞λi(T2)<∞.
Mercer's theorem
edit edit sourceMercer's theorem provides another example of a trace class operator. That is, suppose K is a continuous symmetric positive-definite kernel on L2([a,b]), defined as
- K(s,t)=∑j=1∞λjej(s)ej(t)
then the associated Hilbert–Schmidt integral operator TK is trace class, i.e.,
- Tr(TK)=∫abK(t,t)dt=∑iλi.
Finite-rank operators
edit edit sourceEvery finite-rank operator is a trace-class operator. Furthermore, the space of all finite-rank operators is a dense subspace of B1(H) (when endowed with the trace norm).[9]
Given any x,y∈H, define the operator x⊗y:H→H by (x⊗y)(z):=⟨z,y⟩x. Then x⊗y is a continuous linear operator of rank 1 and is thus trace class; moreover, for any bounded linear operator A on H (and into H), Tr(A(x⊗y))=⟨Ax,y⟩.[9]
Properties
edit edit source- If A:H→H is a non-negative self-adjoint operator, then A is trace-class if and only if TrA<∞. Therefore, a self-adjoint operator A is trace-class if and only if its positive part A+ and negative part A− are both trace-class. (The positive and negative parts of a self-adjoint operator are obtained by the continuous functional calculus.)
- The trace is a linear functional over the space of trace-class operators, that is, Tr(aA+bB)=aTr(A)+bTr(B). The bilinear map ⟨A,B⟩=Tr(A∗B) is an inner product on the trace class; the corresponding norm is called the Hilbert–Schmidt norm. The completion of the trace-class operators in the Hilbert–Schmidt norm are called the Hilbert–Schmidt operators.
- Tr:B1(H)→ℂ is a positive linear functional such that if T is a trace class operator satisfying T≥0 and TrT=0, then T=0.[11]
- If T:H→H is trace-class then so is T∗ and ‖T‖1=‖T∗‖1.[11]
- If A:H→H is bounded, and T:H→H is trace-class, then AT and TA are also trace-class (i.e. the space of trace-class operators on H is a two-sided ideal in the algebra of bounded linear operators on H), and[11][13] ‖AT‖1=Tr(|AT|)≤‖A‖‖T‖1,‖TA‖1=Tr(|TA|)≤‖A‖‖T‖1. Furthermore, under the same hypothesis,[11] Tr(AT)=Tr(TA) and |Tr(AT)|≤‖A‖‖T‖. The last assertion also holds under the weaker hypothesis that A and T are Hilbert–Schmidt.
- If (ek)k and (fk)k are two orthonormal bases of H and if T is trace class then ∑k|⟨Tek,fk⟩|≤‖T‖1.[9]
- If A is trace-class, then one can define the Fredholm determinant of I+A: det(I+A):=∏n≥1[1+λn(A)], where {λn(A)}n is the spectrum of A. The trace class condition on A guarantees that the infinite product is finite: indeed, det(I+A)≤e‖A‖1. It also implies that det(I+A)≠0 if and only if (I+A) is invertible.
- If A:H→H is trace class then for any orthonormal basis (ek)k of H, the sum of positive terms ∑k|⟨Aek,ek⟩| is finite.[11]
- If A=B∗C for some Hilbert-Schmidt operators B and C then for any normal vector e∈H, |⟨Ae,e⟩|≤12(‖Be‖2+‖Ce‖2) holds.[11]
Lidskii's theorem
edit edit sourceLet A be a trace-class operator in a separable Hilbert space H, and let {λn(A)}n=1N≤∞ be the eigenvalues of A. Let us assume that λn(A) are enumerated with algebraic multiplicities taken into account (that is, if the algebraic multiplicity of λ is k, then λ is repeated k times in the list λ1(A),λ2(A),…). Lidskii's theorem (named after Victor Borisovich Lidskii) states that Tr(A)=∑n=1Nλn(A)
Note that the series on the right converges absolutely due to Weyl's inequality ∑n=1N|λn(A)|≤∑m=1Msm(A) between the eigenvalues {λn(A)}n=1N and the singular values {sm(A)}m=1M of the compact operator A.[14] See also Grothendieck trace theorem.
Relationship between common classes of operators
edit edit sourceOne can view certain classes of bounded operators as noncommutative analogue of classical sequence spaces, with trace-class operators as the noncommutative analogue of the sequence space ℓ1(ℕ).
Indeed, it is possible to apply the spectral theorem to show that every normal trace-class operator on a separable Hilbert space can be realized in a certain way as an ℓ1 sequence with respect to some choice of a pair of Hilbert bases. In the same vein, the bounded operators are noncommutative versions of ℓ∞(ℕ), the compact operators that of c0 (the sequences convergent to 0), Hilbert–Schmidt operators correspond to ℓ2(ℕ), and finite-rank operators to c00 (the sequences that have only finitely many non-zero terms). To some extent, the relationships between these classes of operators are similar to the relationships between their commutative counterparts.
Recall that every compact operator T on a Hilbert space takes the following canonical form: there exist orthonormal bases (ui)i and (vi)i and a sequence (αi)i of non-negative numbers with αi→0 such that Tx=∑iαi⟨x,vi⟩ui for all x∈H. Making the above heuristic comments more precise, we have that T is trace-class iff the series ∑iαi is convergent, T is Hilbert–Schmidt iff ∑iαi2 is convergent, and T is finite-rank iff the sequence (αi)i has only finitely many nonzero terms. This allows to relate these classes of operators. The following inclusions hold and are all proper when H is infinite-dimensional:{ finite rank }⊆{ trace class }⊆{ Hilbert--Schmidt }⊆{ compact }.
The trace-class operators are given the trace norm ‖T‖1=Tr[(T∗T)1/2]=∑iαi. The norm corresponding to the Hilbert–Schmidt inner product is ‖T‖2=[Tr(T∗T)]1/2=(∑iαi2)1/2. Also, the usual operator norm is ‖T‖=supi(αi). By classical inequalities regarding sequences, ‖T‖≤‖T‖2≤‖T‖1 for appropriate T.
It is also clear that finite-rank operators are dense in both trace-class and Hilbert–Schmidt in their respective norms.
Trace class as the dual of compact operators
edit edit sourceThe dual space of c0 is ℓ1(ℕ). Similarly, we have that the dual of compact operators, denoted by K(H)∗, is the trace-class operators, denoted by B1. The argument, which we now sketch, is reminiscent of that for the corresponding sequence spaces. Let f∈K(H)∗, we identify f with the operator Tf defined by ⟨Tfx,y⟩=f(Sx,y), where Sx,y is the rank-one operator given by Sx,y(h)=⟨h,y⟩x.
This identification works because the finite-rank operators are norm-dense in K(H). In the event that Tf is a positive operator, for any orthonormal basis ui, one has ∑i⟨Tfui,ui⟩=f(I)≤‖f‖, where I is the identity operator: I=∑i⟨⋅,ui⟩ui.
But this means that Tf is trace-class. An appeal to polar decomposition extend this to the general case, where Tf need not be positive.
A limiting argument using finite-rank operators shows that ‖Tf‖1=‖f‖. Thus K(H)∗ is isometrically isomorphic to B1.
As the predual of bounded operators
edit edit sourceRecall that the dual of ℓ1(ℕ) is ℓ∞(ℕ). In the present context, the dual of trace-class operators B1 is the bounded operators B(H). More precisely, the set B1 is a two-sided ideal in B(H). So given any operator T∈B(H), we may define a continuous linear functional φT on B1 by φT(A)=Tr(AT). This correspondence between bounded linear operators and elements φT of the dual space of B1 is an isometric isomorphism. It follows that B(H) is the dual space of B1. This can be used to define the weak-* topology on B(H).
See also
edit edit sourceReferences
edit edit source- ^ Mittelstaedt 2009, pp. 389–390. Wikipedia revision 1354562143, retrieved 2026-11-18
- ^ a b Conway 2000, p. 86. Wikipedia revision 1354562143, retrieved 2026-11-18
- ^ Reed & Simon 1980, p. 206. Wikipedia revision 1354562143, retrieved 2026-11-18
- ^ Reed & Simon 1980, p. 196. Wikipedia revision 1354562143, retrieved 2026-11-18
- ^ Reed & Simon 1980, p. 195. Wikipedia revision 1354562143, retrieved 2026-11-18
- ^ Trèves 2006, p. 494. Wikipedia revision 1354562143, retrieved 2026-11-18
- ^ Conway 2000, p. 89. Wikipedia revision 1354562143, retrieved 2026-11-18
- ^ Reed & Simon 1980, pp. 203–204, 209. Wikipedia revision 1354562143, retrieved 2026-11-18
- ^ a b c d Conway 1990, p. 268. Wikipedia revision 1354562143, retrieved 2026-11-18
- ^ Trèves 2006, pp. 502–508. Wikipedia revision 1354562143, retrieved 2026-11-18
- ^ a b c d e f g h Conway 1990, p. 267. Wikipedia revision 1354562143, retrieved 2026-11-18
- ^ Simon 2010, p. 21. Wikipedia revision 1354562143, retrieved 2026-11-18
- ^ Reed & Simon 1980, p. 218. Wikipedia revision 1354562143, retrieved 2026-11-18
- ^ Simon, B. (2005) Trace ideals and their applications, Second Edition, American Mathematical Society. Wikipedia revision 1354562143, retrieved 2026-11-18
Bibliography
edit edit source- Conway, John B. (2000). A Course in Operator Theory. Providence (R.I.): American Mathematical Soc. ISBN 978-0-8218-2065-0.
- Conway, John B. (1990). A course in functional analysis. New York: Springer-Verlag. ISBN 978-0-387-97245-9. OCLC 21195908.
- Dixmier, J. (1969). Les Algebres d'Operateurs dans l'Espace Hilbertien. Gauthier-Villars.
- Mittelstaedt, Peter (2009). "Mixed State". Compendium of Quantum Physics. Berlin, Heidelberg: Springer Berlin Heidelberg. pp. 389–390. doi:10.1007/978-3-540-70626-7_120. ISBN 978-3-540-70622-9.
- Reed, M.; Simon, B. (1980). Methods of Modern Mathematical Physics: Vol 1: Functional analysis. Academic Press. ISBN 978-0-12-585050-6.
- Schaefer, Helmut H. (1999). Topological Vector Spaces. GTM. Vol. 3. New York, NY: Springer New York Imprint Springer. ISBN 978-1-4612-7155-0. OCLC 840278135.
- Simon, Barry (2010). Szegő's theorem and its descendants: spectral theory for L² perturbations of orthogonal polynomials. Princeton University Press. ISBN 978-0-691-14704-8.
- Trèves, François (2006) [1967]. Topological Vector Spaces, Distributions and Kernels. Mineola, N.Y.: Dover Publications. ISBN 978-0-486-45352-1. OCLC 853623322.
Text derived from https://en.wikipedia.org/w/index.php?oldid=1354562143 under CC BY-SA 4.0; modifications have been made.