-
The Witten index and the spectral shift function
Authors:
Alan Carey,
Galina Levitina,
Denis Potapov,
Fedor Sukochev
Abstract:
In \cite{APSIII} Atiyah, Patodi and Singer introduced spectral flow for elliptic operators on odd dimensional compact manifolds. They argued that it could be computed from the Fredholm index of an elliptic operator on a manifold of one higher dimension. A general proof of this fact was produced by Robbin-Salamon \cite{RS95}. In \cite{GLMST}, a start was made on extending these ideas to operators w…
▽ More
In \cite{APSIII} Atiyah, Patodi and Singer introduced spectral flow for elliptic operators on odd dimensional compact manifolds. They argued that it could be computed from the Fredholm index of an elliptic operator on a manifold of one higher dimension. A general proof of this fact was produced by Robbin-Salamon \cite{RS95}. In \cite{GLMST}, a start was made on extending these ideas to operators with some essential spectrum as occurs on non-compact manifolds. The new ingredient introduced there was to exploit scattering theory following the fundamental paper \cite{Pu08}. These results do not apply to differential operators directly, only to pseudo-differential operators on manifolds, due to the restrictive assumption that spectral flow is considered between an operator and {its perturbation by a relatively trace-class operator}. In this paper we extend the main results of these earlier papers to spectral flow between an operator and a perturbation satisfying a higher $p^{th}$ Schatten class condition for $0\leq p<\infty$, thus allowing differential operators on manifolds of any dimension $d<p+1$. In fact our main result does not assume any ellipticity or Fredholm properties at all and proves an operator theoretic trace formula motivated by \cite{BCPRSW, CGK16}. We illustrate our results using Dirac type operators on $L^2(\bbR^d)$ for arbitrary $d\in\bbN$. In this setting our main result substantially extends \cite[Theorem 3.5]{CGGLPSZ16}, where the case $d=1$ was treated.
△ Less
Submitted 22 February, 2022; v1 submitted 17 January, 2021;
originally announced January 2021.
-
On a trace formula for functions of noncommuting operators
Authors:
A. B. Aleksandrov,
V. V. Peller,
D. S. Potapov
Abstract:
The main result of the paper is that the Lifshits--Krein trace formula cannot be generalized to the case of functions of noncommuting self-adjoint operators. To prove this, we show that for pairs $(A_1,B_1)$ and $(A_2,B_2)$ of bounded self-adjoint operators with trace class differences $A_2-A_1$ and $B_2-B_1$, it is impossible to estimate the modulus of the trace of the difference…
▽ More
The main result of the paper is that the Lifshits--Krein trace formula cannot be generalized to the case of functions of noncommuting self-adjoint operators. To prove this, we show that for pairs $(A_1,B_1)$ and $(A_2,B_2)$ of bounded self-adjoint operators with trace class differences $A_2-A_1$ and $B_2-B_1$, it is impossible to estimate the modulus of the trace of the difference $f(A_2,B_2)-f(A_1,B_1)$ in terms of the norm of $f$ in the Lipschitz class.
△ Less
Submitted 27 January, 2019;
originally announced January 2019.
-
On the Global Limiting Absorption Principle for Massless Dirac Operators
Authors:
Alan Carey,
Fritz Gesztesy,
Jens Kaad,
Galina Levitina,
Roger Nichols,
Denis Potapov,
Fedor Sukochev
Abstract:
We prove a global limiting absorption principle on the entire real line for free, massless Dirac operators $H_0 = α\cdot (-i \nabla)$ for all space dimensions $n \in \mathbb{N}$, $n \geq 2$. This is a new result for all dimensions other than three, in particular, it applies to the two-dimensional case which is known to be of some relevance in applications to graphene.
We also prove an essential…
▽ More
We prove a global limiting absorption principle on the entire real line for free, massless Dirac operators $H_0 = α\cdot (-i \nabla)$ for all space dimensions $n \in \mathbb{N}$, $n \geq 2$. This is a new result for all dimensions other than three, in particular, it applies to the two-dimensional case which is known to be of some relevance in applications to graphene.
We also prove an essential self-adjointness result for first-order matrix-valued differential operators with Lipschitz coefficients.
△ Less
Submitted 3 November, 2017;
originally announced November 2017.
-
Krein's trace theorem revisited
Authors:
Denis Potapov,
Fedor Sukochev,
Dmitriy Zanin
Abstract:
We supply the first proof of Krein's Trace Theorem which does not use complex analysis. Our proof holds for~$σ$-finite von Neumann algebras $\mathcal{M}$ of type II and unbounded perturbations from the predual of~$\mathcal{M}$.
We supply the first proof of Krein's Trace Theorem which does not use complex analysis. Our proof holds for~$σ$-finite von Neumann algebras $\mathcal{M}$ of type II and unbounded perturbations from the predual of~$\mathcal{M}$.
△ Less
Submitted 14 December, 2016;
originally announced January 2017.
-
Singular traces and perturbation formulae of higher order
Authors:
Denis Potapov,
Fedor Sukochev,
Aleksandr Usachev,
Dmitriy Zanin
Abstract:
Let $H, V$ be self-adjoint operators such that $V$ belongs to the weak trace class ideal. We prove higher order perturbation formula $$τ\big(f(H+V)-\sum_{j=0}^{n-1}\frac{1}{j!}\frac{d^j}{dt^j} f(H+tV)\big|_{t=0}\big)=\int_{\mathbb{R}} f^{(n)}(t)\,dm_n(t),$$ where $τ$ is a trace on the weak trace class ideal and $m_n$ is a finite measure that is not necessarily absolutely continuous. This result ex…
▽ More
Let $H, V$ be self-adjoint operators such that $V$ belongs to the weak trace class ideal. We prove higher order perturbation formula $$τ\big(f(H+V)-\sum_{j=0}^{n-1}\frac{1}{j!}\frac{d^j}{dt^j} f(H+tV)\big|_{t=0}\big)=\int_{\mathbb{R}} f^{(n)}(t)\,dm_n(t),$$ where $τ$ is a trace on the weak trace class ideal and $m_n$ is a finite measure that is not necessarily absolutely continuous. This result extends the first and second order perturbation formulas of Dykema and Shripka, who generalised the Krein and Koplienko trace formulas to the weak trace class ideal. We also establish the perturbation formulae when the perturbation $V$ belongs to the quasi-Banach ideal weak-$L_n$ for any $n \geq 1$.
△ Less
Submitted 14 December, 2016;
originally announced December 2016.
-
Trace Formulas for a Class of non-Fredholm Operators: A Review
Authors:
Alan Carey,
Fritz Gesztesy,
Harald Grosse,
Galina Levitina,
Denis Potapov,
Fedor Sukochev,
Dmitriy Zanin
Abstract:
We review previous work on spectral flow in connection with certain self-adjoint model operators $\{A(t)\}_{t\in \mathbb{R}}$ on a Hilbert space $\mathcal{H}$, joining endpoints $A_\pm$, and the index of the operator $D_{A}^{}= (d/d t) + A$ acting in $L^2(\mathbb{R}; \mathcal{H})$, where $A$ denotes the operator of multiplication $(A f)(t) = A(t)f(t)$. In this article we review what is known when…
▽ More
We review previous work on spectral flow in connection with certain self-adjoint model operators $\{A(t)\}_{t\in \mathbb{R}}$ on a Hilbert space $\mathcal{H}$, joining endpoints $A_\pm$, and the index of the operator $D_{A}^{}= (d/d t) + A$ acting in $L^2(\mathbb{R}; \mathcal{H})$, where $A$ denotes the operator of multiplication $(A f)(t) = A(t)f(t)$. In this article we review what is known when these operators have some essential spectrum and describe some new results in terms of associated spectral shift functions.
We are especially interested in extensions to non-Fredholm situations, replacing the Fredholm index by the Witten index, and use a particular $(1+1)$-dimensional model setup to illustrate our approach based on spectral shift functions.
△ Less
Submitted 16 October, 2016;
originally announced October 2016.
-
Double operator integral methods applied to continuity of spectral shift functions
Authors:
Alan Carey,
Fritz Gesztesy,
Galina Levitina,
Roger Nichols,
Denis Potapov,
Fedor Sukochev
Abstract:
We derive two main results: First, assume that $A$, $B$, $A_n$, $B_n$ are self-adjoint operators in the Hilbert space $\mathcal{H}$, and suppose that $A_n$ converges to $A$ and $B_n$ to $B$ in strong resolvent sense as $n \to \infty$. Fix $m \in \mathbb{N}$, $m$ odd, $p \in [1,\infty)$, and assume that…
▽ More
We derive two main results: First, assume that $A$, $B$, $A_n$, $B_n$ are self-adjoint operators in the Hilbert space $\mathcal{H}$, and suppose that $A_n$ converges to $A$ and $B_n$ to $B$ in strong resolvent sense as $n \to \infty$. Fix $m \in \mathbb{N}$, $m$ odd, $p \in [1,\infty)$, and assume that $T:= \big[( A + iI_{\mathcal{H}})^{-m} - ( B + iI_{\mathcal{H}})^{-m}\big] \in \mathcal{B}_p(\mathcal{H})$, $T_n := \big[( A_n + iI_{\mathcal{H}})^{-m} - ( B_n + iI_{\mathcal{H}})^{-m}\big] \in \mathcal{B}_p(\mathcal{H})$, and $\lim_{n \rightarrow \infty} \|T_n - T\|_{\mathcal{B}_p(\mathcal{H})} =0$. Then for any function $f$ in the class $\mathfrak F_{k}(\mathbb{R}) \supset C_0^{\infty}(\mathbb{R})$ (cf. (1.1)), $$ \lim_{n \rightarrow \infty} \big\| [f(A_n) - f(B_n)] - [f(A)- f(B)]\big\|_{\mathcal{B}_p(\mathcal{H})}=0. $$
Our second result concerns the continuity of spectral shift functions $ξ(\cdot; B,B_0)$ with respect to the operator parameter $B$. For $T$ self-adjoint in $\mathcal{H}$ we denote by $Γ_m(T)$, $m \in \mathbb{N}$ odd, the set of all self-adjoint operators $S$ in $\mathcal{H}$ satisfying $\big[(S - z I_{\mathcal{H}})^{-m} - (T - z I_{\mathcal{H}})^{-m}\big] \in \mathcal{B}_1(\mathcal{H})$, $z \in \mathbb{C}\backslash \mathbb{R}$. Employing a suitable topology on $Γ_m(T)$ (cf. (1.9), we prove the following: Suppose that $B_1\in Γ_m(B_0)$ and let $\{B_τ\}_{τ\in [0,1]}\subset Γ_m(B_0)$ denote a path from $B_0$ to $B_1$ in $Γ_m(B_0)$ depending continuously on $τ\in [0,1]$ with respect to the topology on $Γ_m(B_0)$. If $f \in L^{\infty}(\mathbb{R})$, then $$ \lim_{τ\to 0^+} \|ξ(\, \cdot \, ; B_τ, A_0) f - ξ(\, \cdot \, ; B_0, A_0) f\|_{L^1(\mathbb{R}; (|ν|^{m+1} + 1)^{-1}dν)} = 0. $$
△ Less
Submitted 2 February, 2016; v1 submitted 25 November, 2015;
originally announced November 2015.
-
On Index Theory for Non-Fredholm Operators: A $(1+1)$-Dimensional Example
Authors:
Alan Carey,
Fritz Gesztesy,
Galina Levitina,
Denis Potapov,
Fedor Sukochev,
Dima Zanin
Abstract:
Using the general formalism of [12], a study of index theory for non-Fredholm operators was initiated in [9]. Natural examples arise from $(1+1)$-dimensional differential operators using the model operator $D_A$ in $L^2(\mathbb{R}^2; dt dx)$ of the type $D_A = (d/dt) + A$, where $A = \int^{\oplus}_{\mathbb{R}} dt \, A(t)$, and the family of self-adjoint operators $A(t)$ in $L^2(\mathbb{R}; dx)$ is…
▽ More
Using the general formalism of [12], a study of index theory for non-Fredholm operators was initiated in [9]. Natural examples arise from $(1+1)$-dimensional differential operators using the model operator $D_A$ in $L^2(\mathbb{R}^2; dt dx)$ of the type $D_A = (d/dt) + A$, where $A = \int^{\oplus}_{\mathbb{R}} dt \, A(t)$, and the family of self-adjoint operators $A(t)$ in $L^2(\mathbb{R}; dx)$ is explicitly given by $A(t) = - i (d/dx) + θ(t) φ(\cdot)$, $t \in \mathbb{R}$. Here $φ: \mathbb{R} \to \mathbb{R}$ has to be integrable on $\mathbb{R}$ and $θ: \mathbb{R} \to \mathbb{R}$ tends to zero as $t \to - \infty$ and to $1$ as $t \to + \infty$. In particular, $A(t)$ has asymptotes in the norm resolvent sense $A_- = - i (d/dx)$, $A_+ = - i (d/dx) + φ(\cdot)$ as $t \to \mp \infty$.
Since $D_A$ violates the relative trace class condition introduced in [9], we now employ a new approach based on an approximation technique. The approximants do fit the framework of [9] and lead to the following results: Introducing $H_1 = {D_A}^* D_A$, $H_2 = D_A {D_A}^*$, we recall that the resolvent regularized Witten index of $D_A$, denoted by $W_r(D_A)$, is defined by $$ W_r(D_A) = \lim_{λ\to 0} (- λ) {\rm tr}_{L^2(\mathbb{R}^2; dtdx)}((H_1 - λI)^{-1} - (H_2 - λI)^{-1}). $$ In the concrete example at hand, we prove $$ W_r(D_A) = ξ(0_+; H_2, H_1) = ξ(0; A_+, A_-) = 1/(2 π) \int_{\mathbb{R}} dx \, φ(x). $$ Here $ξ(\, \cdot \, ; S_2, S_1)$, denotes the spectral shift operator for the pair $(S_2,S_1)$, and we employ the normalization, $ξ(λ; H_2, H_1) = 0$, $λ< 0$.
△ Less
Submitted 4 September, 2015;
originally announced September 2015.
-
Peller's problem concerning Koplienko-Neidhardt trace formulae: the unitary case
Authors:
Clément Coine,
Christian Le Merdy,
Denis Potapov,
Fedor Sukochev,
Anna Tomskova
Abstract:
We prove the existence of a complex valued $C^2$-function on the unit circle, a unitary operator U and a self-adjoint operator Z in the Hilbert-Schmidt class $S^2$, such that the perturbated operator $$ f(e^{iZ}U)-f(U) -\frac{d}{dt}\bigl(f(e^{itZ}U)\bigr)_{\vert t=0} $$ does not belong to the space $S^1$ of trace class operators. This resolves a problem of Peller concerning the validity of the Kop…
▽ More
We prove the existence of a complex valued $C^2$-function on the unit circle, a unitary operator U and a self-adjoint operator Z in the Hilbert-Schmidt class $S^2$, such that the perturbated operator $$ f(e^{iZ}U)-f(U) -\frac{d}{dt}\bigl(f(e^{itZ}U)\bigr)_{\vert t=0} $$ does not belong to the space $S^1$ of trace class operators. This resolves a problem of Peller concerning the validity of the Koplienko-Neidhardt trace formula for unitaries.
△ Less
Submitted 2 September, 2015;
originally announced September 2015.
-
Beat-Event Detection in Action Movie Franchises
Authors:
Danila Potapov,
Matthijs Douze,
Jerome Revaud,
Zaid Harchaoui,
Cordelia Schmid
Abstract:
While important advances were recently made towards temporally localizing and recognizing specific human actions or activities in videos, efficient detection and classification of long video chunks belonging to semantically defined categories such as "pursuit" or "romance" remains challenging.We introduce a new dataset, Action Movie Franchises, consisting of a collection of Hollywood action movie…
▽ More
While important advances were recently made towards temporally localizing and recognizing specific human actions or activities in videos, efficient detection and classification of long video chunks belonging to semantically defined categories such as "pursuit" or "romance" remains challenging.We introduce a new dataset, Action Movie Franchises, consisting of a collection of Hollywood action movie franchises. We define 11 non-exclusive semantic categories - called beat-categories - that are broad enough to cover most of the movie footage. The corresponding beat-events are annotated as groups of video shots, possibly overlap**.We propose an approach for localizing beat-events based on classifying shots into beat-categories and learning the temporal constraints between shots. We show that temporal constraints significantly improve the classification performance. We set up an evaluation protocol for beat-event localization as well as for shot classification, depending on whether movies from the same franchise are present or not in the training data.
△ Less
Submitted 15 August, 2015;
originally announced August 2015.
-
Weak type commutator and Lipschitz estimates: resolution of the Nazarov-Peller conjecture
Authors:
Martijn Caspers,
Denis Potapov,
Fedor Sukochev,
Dmitriy Zanin
Abstract:
Let $\mathcal{M}$ be a semi-finite von Neumann algebra and let $f: \mathbb{R} \rightarrow \mathbb{C}$ be a Lipschitz function. If $A,B\in\mathcal{M}$ are self-adjoint operators such that $[A,B]\in L_1(\mathcal{M}),$ then $$\|[f(A),B]\|_{1,\infty}\leq c_{abs}\|f'\|_{\infty}\|[A,B]\|_1,$$ where $c_{abs}$ is an absolute constant independent of $f$, $\mathcal{M}$ and $A,B$ and $\|\cdot\|_{1,\infty}$ d…
▽ More
Let $\mathcal{M}$ be a semi-finite von Neumann algebra and let $f: \mathbb{R} \rightarrow \mathbb{C}$ be a Lipschitz function. If $A,B\in\mathcal{M}$ are self-adjoint operators such that $[A,B]\in L_1(\mathcal{M}),$ then $$\|[f(A),B]\|_{1,\infty}\leq c_{abs}\|f'\|_{\infty}\|[A,B]\|_1,$$ where $c_{abs}$ is an absolute constant independent of $f$, $\mathcal{M}$ and $A,B$ and $\|\cdot\|_{1,\infty}$ denotes the weak $L_1$-norm. If $X,Y\in\mathcal{M}$ are self-adjoint operators such that $X-Y\in L_1(\mathcal{M}),$ then $$\|f(X)-f(Y)\|_{1,\infty}\leq c_{abs}\|f'\|_{\infty}\|X-Y\|_1.$$ This result resolves a conjecture raised by F. Nazarov and V. Peller implying a couple of existing results in perturbation theory.
△ Less
Submitted 2 June, 2015;
originally announced June 2015.
-
Resolution of Peller's problem concerning Koplienko-Neidhardt trace formulae
Authors:
Clément Coine,
Christian Le Merdy,
Denis Potapov,
Fedor Sukochev,
Anna Tomskova
Abstract:
A formula for the norm of a bilinear Schur multiplier acting from the Cartesian product $\mathcal S^2\times \mathcal S^2$ of two copies of the Hilbert-Schmidt classes into the trace class $\mathcal S^1$ is established in terms of linear Schur multipliers acting on the space $\mathcal S^\infty$ of all compact operators. Using this formula, we resolve Peller's problem on Koplienko-Neidhardt trace fo…
▽ More
A formula for the norm of a bilinear Schur multiplier acting from the Cartesian product $\mathcal S^2\times \mathcal S^2$ of two copies of the Hilbert-Schmidt classes into the trace class $\mathcal S^1$ is established in terms of linear Schur multipliers acting on the space $\mathcal S^\infty$ of all compact operators. Using this formula, we resolve Peller's problem on Koplienko-Neidhardt trace formulae. Namely, we prove that there exist a twice continuously differentiable function $f$ with a bounded second derivative, a self-adjoint (unbounded) operator $A$ and a self-adjoint operator $B\in \mathcal S^2$ such that $$ f(A+B)-f(A)-\frac{d}{dt}(f(A+tB))\big\vert_{t=0}\notin \mathcal S^1. $$
△ Less
Submitted 15 April, 2015;
originally announced April 2015.
-
On the Witten index in terms of spectral shift functions
Authors:
Alan Carey,
Fritz Gesztesy,
Denis Potapov,
Fedor Sukochev,
Yuri Tomilov
Abstract:
We study the model operator $\mathbf{D}_{\mathbf{A}} = (d/dt) + \mathbf{A}$ in $L^2(\mathbb{R};\mathcal{H})$ associated with the operator path $\{A(t)\}_{t=-\infty}^{\infty}$, where $(\mathbf{A} f)(t) = A(t) f(t)$ for a.e.\ $t\in\mathbb{R}$, and appropriate $f \in L^2(\mathbb{R};\mathcal{H})$ (with $\mathcal{H}$ a separable, complex Hilbert space). Denoting by $A_{\pm}$ the norm resolvent limits o…
▽ More
We study the model operator $\mathbf{D}_{\mathbf{A}} = (d/dt) + \mathbf{A}$ in $L^2(\mathbb{R};\mathcal{H})$ associated with the operator path $\{A(t)\}_{t=-\infty}^{\infty}$, where $(\mathbf{A} f)(t) = A(t) f(t)$ for a.e.\ $t\in\mathbb{R}$, and appropriate $f \in L^2(\mathbb{R};\mathcal{H})$ (with $\mathcal{H}$ a separable, complex Hilbert space). Denoting by $A_{\pm}$ the norm resolvent limits of $A(t)$ as $t \to \pm \infty$, our setup permits $A(t)$ in $\mathcal{H}$ to be an unbounded, relatively trace class perturbation of the unbounded self-adjoint operator $A_-$, and no discrete spectrum assumptions are made on $A_{\pm}$.
We introduce resolvent and semigroup regularized Witten indices of $\mathbf{D}_{\mathbf{A}}$, denoted by $W_r$ and $W_s$, and prove that these regularized indices coincide with the Fredholm index of $\mathbf{D}_{\mathbf{A}}$ whenever the latter is Fredholm.
In situations where $\mathbf{D}_{\mathbf{A}}$ ceases to be a Fredholm operator in $L^2(\mathbb{R};\mathcal{H})$ we compute its resolvent (resp., semigroup) regularized Witten index in terms of the spectral shift function $ξ(\,\cdot\,;A_+,A_-)$ associated with the pair $(A_+, A_-)$ as follows: Assuming $0$ to be a right and a left Lebesgue point of $ξ(\,\cdot\,\, ; A_+, A_-)$, denoted by $ξ_L(0_+; A_+,A_-)$ and $ξ_L(0_-; A_+, A_-)$, we prove that $0$ is also a right Lebesgue point of $ξ(\,\cdot\,\, ; \mathbf{H_2}, \mathbf{H_1})$, denoted by $ξ_L(0_+; \mathbf{H_2}, \mathbf{H_1})$, and that \begin{align*} W_r(\mathbf{D}_{\mathbf{A}}) &= W_s(\mathbf{D}_{\mathbf{A}}) \\ & = ξ_L(0_+; \mathbf{H_2}, \mathbf{H_1}) \\ & = [ξ_L(0_+; A_+,A_-) + ξ_L(0_-; A_+, A_-)]/2, \end{align*} the principal result of this paper.
In the special case where $\dim(\mathcal{H}) < \infty$, we prove that the Witten indices of $\mathbf{D}_{\mathbf{A}}$ are either integer, or half-integer-valued.
△ Less
Submitted 10 September, 2014; v1 submitted 2 April, 2014;
originally announced April 2014.
-
A Jost-Pais-type reduction of Fredholm determinants and some applications
Authors:
Alan Carey,
Fritz Gesztesy,
Denis Potapov,
Fedor Sukochev,
Yuri Tomilov
Abstract:
We study the analog of semi-separable integral kernels in $\cH$ of the type {equation*} K(x,x')={cases} F_1(x)G_1(x'), & a<x'< x< b, \\ F_2(x)G_2(x'), & a<x<x'<b, {cases} {equation*} where $-\infty\leq a<b\leq \infty$, and for a.e.\ $x \in (a,b)$, $F_j (x) \in \cB_2(\cH_j,\cH)$ and $G_j(x) \in \cB_2(\cH,\cH_j)$ such that $F_j(\cdot)$ and $G_j(\cdot)$ are uniformly measurable, and {equation*} \|F_j…
▽ More
We study the analog of semi-separable integral kernels in $\cH$ of the type {equation*} K(x,x')={cases} F_1(x)G_1(x'), & a<x'< x< b, \\ F_2(x)G_2(x'), & a<x<x'<b, {cases} {equation*} where $-\infty\leq a<b\leq \infty$, and for a.e.\ $x \in (a,b)$, $F_j (x) \in \cB_2(\cH_j,\cH)$ and $G_j(x) \in \cB_2(\cH,\cH_j)$ such that $F_j(\cdot)$ and $G_j(\cdot)$ are uniformly measurable, and {equation*} \|F_j(\cdot)\|_{\cB_2(\cH_j,\cH)} \in L^2((a,b)), \; \|G_j (\cdot)\|_{\cB_2(\cH,\cH_j)} \in L^2((a,b)), \quad j=1,2, {equation*} with $\cH$ and $\cH_j$, $j=1,2$, complex, separable Hilbert spaces. Assuming that $K(\cdot, \cdot)$ generates a trace class operator $\bsK$ in $L^2((a,b);\cH)$, we derive the analog of the Jost-Pais reduction theory that succeeds in proving that the Fredholm determinant ${\det}_{L^2((a,b);\cH)}(\bsI - α\bsK)$, $α\in \bbC$, naturally reduces to appropriate Fredholm determinants in the Hilbert spaces $\cH$ (and $\cH_1 \oplus \cH_2$).
Explicit applications of this reduction theory are made to Schrödinger operators with suitable bounded operator-valued potentials. In addition, we provide an alternative approach to a fundamental trace formula first established by Pushnitski which leads to a Fredholm index computation of a certain model operator.
△ Less
Submitted 22 April, 2014; v1 submitted 2 April, 2014;
originally announced April 2014.
-
Weak type estimates for the absolute value map**
Authors:
M. Caspers,
D. Potapov,
F. Sukochev,
D. Zanin
Abstract:
We prove that if A and B are bounded self-adjoint operators such that A-B belongs to the trace class, then |A| -|B| belongs to the principal ideal L_{1,\infty} in the algebra L(H) of all bounded operators on an infinite-dimensional Hilbert space generated by an operator whose sequence of eigenvalues is {1, 1/2, 1/3, 1/4, ...}. Moreover, μ(j;|A| -|B|)\leq const(1 + j)^{-1}\|A-B\|_1. We also obtain…
▽ More
We prove that if A and B are bounded self-adjoint operators such that A-B belongs to the trace class, then |A| -|B| belongs to the principal ideal L_{1,\infty} in the algebra L(H) of all bounded operators on an infinite-dimensional Hilbert space generated by an operator whose sequence of eigenvalues is {1, 1/2, 1/3, 1/4, ...}. Moreover, μ(j;|A| -|B|)\leq const(1 + j)^{-1}\|A-B\|_1. We also obtain a semifinite version of this result, as well as the corresponding commutator estimates.
△ Less
Submitted 18 June, 2014; v1 submitted 13 September, 2013;
originally announced September 2013.
-
Frechet differentiability of Sp norms
Authors:
Denis Potapov,
Fedor Sukochev
Abstract:
One of the long standing questions in the theory of Schatten-von Neumann ideals of compact operators is whether their norms have the same differentiability properties as the norms of their commutative counterparts. We answer this question in the affirmative. A key technical observation underlying our proof is a discovery of connection between this question and recent affirmative resolution of L.S.…
▽ More
One of the long standing questions in the theory of Schatten-von Neumann ideals of compact operators is whether their norms have the same differentiability properties as the norms of their commutative counterparts. We answer this question in the affirmative. A key technical observation underlying our proof is a discovery of connection between this question and recent affirmative resolution of L.S. Koplienko's conjecture concerning existence of higher order spectral shift functions.
△ Less
Submitted 3 June, 2013;
originally announced June 2013.
-
Higher order spectral shift for contractions
Authors:
Denis Potapov,
Anna Skripka,
Fedor Sukochev
Abstract:
We derive strong estimates for Schatten norms of operator derivatives along paths of contractions and apply them to prove existence of higher order spectral shift functions for pairs of contractions.
We derive strong estimates for Schatten norms of operator derivatives along paths of contractions and apply them to prove existence of higher order spectral shift functions for pairs of contractions.
△ Less
Submitted 31 October, 2012;
originally announced October 2012.
-
Traces of compact operators and the noncommutative residue
Authors:
Nigel Kalton,
Steven Lord,
Denis Potapov,
Fedor Sukochev
Abstract:
We extend the noncommutative residue of M. Wodzicki on compactly supported classical pseudo-differential operators of order $-d$ and generalise A. Connes' trace theorem, which states that the residue can be calculated using a singular trace on compact operators. Contrary to the role of the noncommutative residue for the classical pseudo-differential operators, a corollary is that the pseudo-differ…
▽ More
We extend the noncommutative residue of M. Wodzicki on compactly supported classical pseudo-differential operators of order $-d$ and generalise A. Connes' trace theorem, which states that the residue can be calculated using a singular trace on compact operators. Contrary to the role of the noncommutative residue for the classical pseudo-differential operators, a corollary is that the pseudo-differential operators of order $-d$ do not have a `unique' trace; pseudo-differential operators can be non-measurable in Connes' sense. Other corollaries are given clarifying the role of Dixmier traces in noncommutative geometry à la Connes, including the definitive statement of Connes' original theorem.
△ Less
Submitted 18 October, 2012; v1 submitted 11 October, 2012;
originally announced October 2012.
-
The best constants for operator Lipschitz functions on Schatten classes
Authors:
Martijn Caspers,
Stephen Montgomery-Smith,
Denis Potapov,
Fedor Sukochev
Abstract:
Suppose that f is a Lipschitz function on the real numbers with Lipschitz constant smaller or equal to 1. Let A be a bounded self-adjoint operator on a Hilbert space H. Let 1<p<infinity and suppose that x in B(H) is an operator such that the commutator [A, x] is contained in the Schatten class S_p. It is proved by the last two authors, that then also [f(A), x] is contained in S_p and there exists…
▽ More
Suppose that f is a Lipschitz function on the real numbers with Lipschitz constant smaller or equal to 1. Let A be a bounded self-adjoint operator on a Hilbert space H. Let 1<p<infinity and suppose that x in B(H) is an operator such that the commutator [A, x] is contained in the Schatten class S_p. It is proved by the last two authors, that then also [f(A), x] is contained in S_p and there exists a constant C_p independent of x and f such that || [f(A), x] ||_p <= C_p || [A,x] ||_p.
The main result of this paper is to give a sharp estimate for C_p in terms of p. Namely, we show that C_p ~ p^2/(p-1). In particular, this gives the best estimates for operator Lipschitz inequalities.
We treat this result in a more general setting. This involves commutators of n self-adjoint operators, for which we prove the analogous result. The case described here in the abstract follows as a special case.
△ Less
Submitted 28 August, 2014; v1 submitted 18 September, 2012;
originally announced September 2012.
-
The Walsh basis in the Lp-spaces of hyperfinite III_λ-factors, 0 < λ<= 1
Authors:
M. Caspers,
D. Potapov,
F. Sukochev
Abstract:
We introduce a non-commutative Walsh system and prove that it forms a Schauder basis in the Lp-spaces (1 < p < \infty) associated with the hyperfinite III_λ-factors (0 < λ<= 1).
We introduce a non-commutative Walsh system and prove that it forms a Schauder basis in the Lp-spaces (1 < p < \infty) associated with the hyperfinite III_λ-factors (0 < λ<= 1).
△ Less
Submitted 10 November, 2011;
originally announced November 2011.
-
On the vector-valued Littlewood-Paley-Rubio de Francia inequality
Authors:
Denis Potapov,
Fedor Sukochev,
Quanhua Xu
Abstract:
The paper studies Banach spaces satisfying the Littlewood-Paley-Rubio de Francia property LPR_p, 2 \leq p < \infty. The paper shows that every Banach lattice whose 2-concavification is a UMD Banach lattice has this property. The paper also shows that every space having LPR_q also has LPR_p with q \leq p < \infty.
The paper studies Banach spaces satisfying the Littlewood-Paley-Rubio de Francia property LPR_p, 2 \leq p < \infty. The paper shows that every Banach lattice whose 2-concavification is a UMD Banach lattice has this property. The paper also shows that every space having LPR_q also has LPR_p with q \leq p < \infty.
△ Less
Submitted 14 April, 2011;
originally announced April 2011.
-
On operator valued Hardy spaces
Authors:
Denis Potapov
Abstract:
The note shows that the operator-valued Hardy space $\sH^1$ introduced via Littlewood-Paley $g$-function coincides with the space of $H^1_R(\T, \sL^1)$ of all Bochner integrable operator-valued functions with integrable analytic part. The proof is based on the noncommutative maximal inequality for Poisson group.
The note shows that the operator-valued Hardy space $\sH^1$ introduced via Littlewood-Paley $g$-function coincides with the space of $H^1_R(\T, \sL^1)$ of all Bochner integrable operator-valued functions with integrable analytic part. The proof is based on the noncommutative maximal inequality for Poisson group.
△ Less
Submitted 7 December, 2010; v1 submitted 29 November, 2010;
originally announced November 2010.
-
Functions of normal operators under perturbations
Authors:
Alexei Aleksandrov,
Vladimir Peller,
Denis Potapov,
Fedor Sukochev
Abstract:
In \cite{Pe1}, \cite{Pe2}, \cite{AP1}, \cite{AP2}, and \cite{AP3} sharp estimates for $f(A)-f(B)$ were obtained for self-adjoint operators $A$ and $B$ and for various classes of functions $f$ on the real line $\R$. In this paper we extend those results to the case of functions of normal operators. We show that if a function $f$ belongs to the Hölder class $Ł_\a(\R^2)$, $0<\a<1$, of functions of tw…
▽ More
In \cite{Pe1}, \cite{Pe2}, \cite{AP1}, \cite{AP2}, and \cite{AP3} sharp estimates for $f(A)-f(B)$ were obtained for self-adjoint operators $A$ and $B$ and for various classes of functions $f$ on the real line $\R$. In this paper we extend those results to the case of functions of normal operators. We show that if a function $f$ belongs to the Hölder class $Ł_\a(\R^2)$, $0<\a<1$, of functions of two variables, and $N_1$ and $N_2$ are normal operators, then $\|f(N_1)-f(N_2)\|\le\const\|f\|_{Ł_\a}\|N_1-N_2\|^\a$. We obtain a more general result for functions in the space $Ł_ø(\R^2)=\big\{f:~|f(\z_1)-f(\z_2)|\le\constø(|\z_1-\z_2|)\big\}$ for an arbitrary modulus of continuity $ø$. We prove that if $f$ belongs to the Besov class $B_{\be1}^1(\R^2)$, then it is operator Lipschitz, i.e., $\|f(N_1)-f(N_2)\|\le\const\|f\|_{B_{\be1}^1}\|N_1-N_2\|$. We also study properties of $f(N_1)-f(N_2)$ in the case when $f\inŁ_\a(\R^2)$ and $N_1-N_2$ belongs to the Schatten-von Neuman class $\bS_p$.
△ Less
Submitted 10 August, 2010;
originally announced August 2010.
-
Functions of perturbed normal operators
Authors:
Aleksei Aleksandrov,
Vladimir Peller,
Denis Potapov,
Fedor Sukochev
Abstract:
In \cite{Pe1}, \cite{Pe2}, \cite{AP1}, \cite{AP2}, and \cite{AP3} sharp estimates for $f(A)-f(B)$ were obtained for self-adjoint operators $A$ and $B$ and for various classes of functions $f$ on the real line $\R$. In this note we extend those results to the case of functions of normal operators. We show that if $f$ belongs to the Hölder class $Ł_\a(\R^2)$, $0<\a<1$, of functions of two variables,…
▽ More
In \cite{Pe1}, \cite{Pe2}, \cite{AP1}, \cite{AP2}, and \cite{AP3} sharp estimates for $f(A)-f(B)$ were obtained for self-adjoint operators $A$ and $B$ and for various classes of functions $f$ on the real line $\R$. In this note we extend those results to the case of functions of normal operators. We show that if $f$ belongs to the Hölder class $Ł_\a(\R^2)$, $0<\a<1$, of functions of two variables, and $N_1$ and $N_2$ are normal operators, then $\|f(N_1)-f(N_2)\|\le\const\|f\|_{Ł_\a}\|N_1-N_2\|^\a$. We obtain a more general result for functions in the space $Ł_ø(\R^2)=\big\{f: |f(\z_1)-f(\z_2)|\le\constø(|\z_1-\z_2|)\big\}$ for an arbitrary modulus of continuity $ø$. We prove that if $f$ belongs to the Besov class $B_{\be1}^1(\R^2)$, then it is operator Lipschitz, i.e., $\|f(N_1)-f(N_2)\|\le\const\|f\|_{B_{\be1}^1}\|N_1-N_2\|$. We also study properties of $f(N_1)-f(N_2)$ in the case when $f\inŁ_\a(\R^2)$ and $N_1-N_2$ belongs to the Schatten-von Neuman class $\bS_p$.
△ Less
Submitted 27 March, 2010;
originally announced March 2010.
-
Spectral shift function of higher order
Authors:
Denis Potapov,
Anna Skripka,
Fedor Sukochev
Abstract:
This paper resolves affirmatively Koplienko's conjecture of 1984 on existence of higher order spectral shift measures. Moreover, the paper establishes absolute continuity of these measures and, thus, existence of the higher order spectral shift functions $η_n$. We show the higher order spectral shift function is a $L^1$-function and prove an estimate on its $L^1$-norm.
Existence and summability…
▽ More
This paper resolves affirmatively Koplienko's conjecture of 1984 on existence of higher order spectral shift measures. Moreover, the paper establishes absolute continuity of these measures and, thus, existence of the higher order spectral shift functions $η_n$. We show the higher order spectral shift function is a $L^1$-function and prove an estimate on its $L^1$-norm.
Existence and summability of $η_1$ and $η_2$ were established by Krein in 1953 and Koplienko in 1984, respectively, whereas for $n > 2$ the problem was unresolved. Our method is derived from [arXiv:0904.4095]; it also applies to the general semi-finite von Neumann algebra setting of the perturbation theory.
△ Less
Submitted 5 November, 2012; v1 submitted 16 December, 2009;
originally announced December 2009.
-
Measures from Dixmier Traces and Zeta Functions
Authors:
Steven Lord,
Denis Potapov,
Fedor Sukochev
Abstract:
For essentially bounded functions on a (closed) compact Riemannian manifold, the noncommutative residue and the Dixmier trace formulation of the noncommutative integral are shown to equate to a multiple of the Lebesgue integral. The identifications are shown to continue to, and be sharp at, square integrable functions. To do better than square integrable, symmetrised noncommutative residue and D…
▽ More
For essentially bounded functions on a (closed) compact Riemannian manifold, the noncommutative residue and the Dixmier trace formulation of the noncommutative integral are shown to equate to a multiple of the Lebesgue integral. The identifications are shown to continue to, and be sharp at, square integrable functions. To do better than square integrable, symmetrised noncommutative residue and Dixmier trace formulas are introduced, for which the identifications are shown to continue to $L^{1+ε}$-spaces, $ε> 0$. However, a failure is shown for the Dixmier trace formulation at integrable functions. The (symmetrised) noncommutative residue and Dixmier trace formulas diverge at this point. It is shown the noncommutative residue remains finite and recovers the Lebesgue integral for any integrable function while the Dixmier trace expression can diverge.
The results show the claim (in the monograph "Elements of Noncommutative Geometry", Birkhauser, 2001), that the identification on smooth functions obtained using Connes' Trace Theorem can be extended to any integrable function, is false.
△ Less
Submitted 8 April, 2010; v1 submitted 8 May, 2009;
originally announced May 2009.
-
Operator-Lipschitz functions in Schatten-von Neumann classes
Authors:
Denis Potapov,
Fedor Sukochev
Abstract:
This paper resolves a number of conjectures in the perturbation theory of linear operators. Namely, we prove that every Lipschitz function is operator Lipschitz in the Schatten-von Neumann ideals $S^α$, $1 < α< \infty$. The negative result for $S^α$, $α= 1, \infty$ was earlier established by Yu. Farforovskaya in 1972.
This paper resolves a number of conjectures in the perturbation theory of linear operators. Namely, we prove that every Lipschitz function is operator Lipschitz in the Schatten-von Neumann ideals $S^α$, $1 < α< \infty$. The negative result for $S^α$, $α= 1, \infty$ was earlier established by Yu. Farforovskaya in 1972.
△ Less
Submitted 14 December, 2009; v1 submitted 27 April, 2009;
originally announced April 2009.
-
Non- (quantum) differentiable $C^1$-functions in the spaces with trivial Boyd indices
Authors:
Denis Potapov,
Fyodor Sukochev
Abstract:
If E is a separable symmetric sequence space with trivial Boyd indices and $\cC^E$ is the corresponding ideal of compact operators, then there exists a $C^1$-function $f_E$, a self-adjoint element $W\in \cC^E$ and a densely defined closed symmetric derivation $δ$ on $\cC^E$ such that $W \in Dom δ$, but $f_E(W) \notin Dom δ$.
If E is a separable symmetric sequence space with trivial Boyd indices and $\cC^E$ is the corresponding ideal of compact operators, then there exists a $C^1$-function $f_E$, a self-adjoint element $W\in \cC^E$ and a densely defined closed symmetric derivation $δ$ on $\cC^E$ such that $W \in Dom δ$, but $f_E(W) \notin Dom δ$.
△ Less
Submitted 20 August, 2008;
originally announced August 2008.
-
Unbounded Fredholm modules and double operator integrals
Authors:
Denis Potapov,
Fyodor Sukochev
Abstract:
In noncommutative geometry one is interested in invariants such as the Fredholm index or spectral flow and their calculation using cyclic cocycles. A variety of formulae have been established under side conditions called summability constraints. These can be formulated in two ways, either for spectral triples or for bounded Fredholm modules. We study the relationship between these by proving var…
▽ More
In noncommutative geometry one is interested in invariants such as the Fredholm index or spectral flow and their calculation using cyclic cocycles. A variety of formulae have been established under side conditions called summability constraints. These can be formulated in two ways, either for spectral triples or for bounded Fredholm modules. We study the relationship between these by proving various properties of the map on unbounded self adjoint operators $D$ given by $f(D)=D(1+D^2)^{-1/2}$. In particular we prove commutator estimates which are needed for the bounded case. In fact our methods work in the setting of semifinite noncommutative geometry where one has $D$ as an unbounded self adjoint linear operator affiliated with a semi-finite von Neumann algebra $\aM$. More precisely we show that for a pair $D,D_0$ of such operators with $D-D_0$ a bounded self-adjoint linear operator from $\aM$ and $ ({\bf 1}+D_0^2)^{-1/2}\in \sE$, where $\sE$ is a noncommutative symmetric space associated with $\aM$, then $$ \Vert f(D) - f (D_0) \Vert_{\sE} \leq C\cdot \Vert D-D_0\Vert_{\aM}. $$ This result is further used to show continuous differentiability of the map** between an odd $\sE$-summable spectral triple and its bounded counterpart.
△ Less
Submitted 20 August, 2008;
originally announced August 2008.
-
The Haar system in the preduals of hyperfinite factors
Authors:
Denis Potapov,
Fyodor Sukochev
Abstract:
We shall present examples of Schauder bases in the preduals to the hyperfinite factors of types $\hbox{II}_1$, $\hbox{II}_\infty$, $\hbox{III}_λ$, $0 < λ\leq 1$. In the semifinite (respectively, purely infinite) setting, these systems form Schauder bases in any associated separable symmetric space of measurable operators (respectively, in any non-commutative $L^p$-space).
We shall present examples of Schauder bases in the preduals to the hyperfinite factors of types $\hbox{II}_1$, $\hbox{II}_\infty$, $\hbox{III}_λ$, $0 < λ\leq 1$. In the semifinite (respectively, purely infinite) setting, these systems form Schauder bases in any associated separable symmetric space of measurable operators (respectively, in any non-commutative $L^p$-space).
△ Less
Submitted 20 August, 2008;
originally announced August 2008.
-
Spectral flow is the integral of one forms on the Banach manifold of self adjoint Fredholm operators
Authors:
Alan Carey,
Denis Potapov,
Fyodor Sukochev
Abstract:
One may trace the idea that spectral flow should be given as the integral of a one form back to the 1974 Vancouver ICM address of I.M. Singer. Our main theorem gives analytic formulae for the spectral flow along a norm differentiable path of self-adjoint bounded Breuer-Fredholm operators in a semi-finite von Neumann algebra. These formulae have a geometric interpretation which derives from the p…
▽ More
One may trace the idea that spectral flow should be given as the integral of a one form back to the 1974 Vancouver ICM address of I.M. Singer. Our main theorem gives analytic formulae for the spectral flow along a norm differentiable path of self-adjoint bounded Breuer-Fredholm operators in a semi-finite von Neumann algebra. These formulae have a geometric interpretation which derives from the proof. Namely we define a family of Banach submanifolds of all bounded self-adjoint Breuer-Fredholm operators and on each submanifold define global one forms whose integral on a norm differentiable path contained in the submanifold calculates the spectral flow along this path. We emphasise that our methods do not give a single globally defined one form on the self adjoint Breuer- Fredholms whose integral along all paths is spectral flow rather, as the choice of the plural `forms' in the title suggests, we need a family of such one forms in order to confirm Singer's idea. The original context for this result concerned paths of unbounded self-adjoint Fredholm operators. We therefore prove analogous formulae for spectral flow in the unbounded case as well. The proof is a synthesis of key contributions by previous authors, whom we acknowledge in detail in the introduction, combined with an additional important recent advance in the differential calculus of functions of non-commuting operators.
△ Less
Submitted 4 June, 2009; v1 submitted 14 July, 2008;
originally announced July 2008.
