BZPEERPapers, people, evidence and research pathsDiscover
Research tools
Network
Opportunities
Account
Source author record
Gian Maria Dall'Ara appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.
Catalog footprint
Research graph
Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.
BZPEER is loading the nearby papers, people, topics and institutions for this page.
Published work
We prove that Heisenberg groups, a.k.a. the boundaries of Siegel domains, minimize the $L^p$ operator norm of the Szegő projection in a large class of weighted CR manifolds of hypersurface type.
In a previous work we proved a spectral multiplier theorem of Mihlin--Hörmander type for two-dimensional Grushin operators $-\partial_x^2 - V(x) \partial_y^2$, where $V$ is a doubling single-well potential, yielding the surprising result that the optimal smoothness requirement on the multiplier is independent of $V$. Here we refine this result, by replacing the $L^\infty$ Sobolev condition on the multiplier with a sharper $L^2$ condition. As a consequence, we obtain the sharp range of $L^1$ boundedness for the associated Bochner--Riesz means. The key new ingredient of the proof is a precise pointwise estimate in the transition region for eigenfunctions of one-dimensional Schrödinger operators with doubling single-well potentials.
The weighted Kohn Laplacian $\Box_φ$ is a natural second order elliptic operator associated to a weight $φ:\mathbb{C}^n\rightarrow\mathbb{R}$ and acting on $(0,1)$-forms, which plays a key role in several questions of complex analysis. We consider here the case of model monomial weights in $\mathbb{C}^2$, i.e., $ φ(z,w):=\sum_{(α,β)\inΓ}|z^αw^β|^2, $ where $Γ\subseteq \mathbb{N}^2$ is finite. Our goal is to prove coercivity estimates of the form $\Box_φ\geq μ^2$, where $μ:\mathbb{C}^n\rightarrow\mathbb{R}$ acts by pointwise multiplication on $(0,1)$-forms, and the inequality is in the sense of self-adjoint operators. We recently proved (arxiv.org:1502.00865) how to derive from $μ$-coercivity estimates for $\Box_φ$ pointwise bounds for the weighted Bergman kernel associated to $φ$. Here we introduce a technique to establish $μ$-coercivity with $ μ(z,w)=c(1+|z|^a+|w|^b) \qquad(a,b\geq0),$ where $a,b\geq0$ depend (and are easily computable from) $Γ$. As a corollary we also prove that, for a wide class of model monomial weights, the spectrum of $\Box_φ$ is discrete if and only if the weight is not decoupled, i.e. $Γ$ contains at least a point $(α,β)$ with $α\neq0\neqβ$. Our methods comprise a new holomorphic uncertainty principle and linear optimization arguments.
We prove three results giving sufficient and/or necessary conditions for discreteness of the spectrum of Schrödinger operators with non-negative matrix-valued potentials, i.e., operators acting on $ψ\in L^2(\mathbb{R}^n,\mathbb{C}^d)$ by the formula $H_Vψ:=-Δψ+Vψ$, where the potential $V$ takes values in the set of non-negative Hermitian $d\times d$ matrices. The first theorem provides a characterization of discreteness of the spectrum when the potential $V$ is in a matrix-valued $A_\infty$ class, thus extending a known result in the scalar case ($d=1$). We also discuss a subtlety in the definition of the appropriate matrix-valued $A_\infty$ class. The second result is a sufficient condition for discreteness of the spectrum, which allows certain degenerate potentials, i.e., such that $\det(V)\equiv0$. To formulate the condition, we introduce a notion of oscillation for subspace-valued mappings. Our third and last result shows that if $V$ is a $2\times2$ real polynomial potential, then $-Δ+V$ has discrete spectrum if and only if the scalar operator $-Δ+λ$ has discrete spectrum, where $λ(x)$ is the minimal eigenvalue of $V(x)$.
The aim of the present thesis is twofold: to study the problem of discreteness of the spectrum of Schrödinger operators with matrix-valued potentials in ${\mathbb R}^d$ (Chapter 1), and to prove new pointwise bounds for weighted Bergman kernels in ${\mathbb C}^n$ (chapters 2 and 3). These two themes are connected by the observation that the weighted Bergman kernel may be effectively studied by means of the analysis of the weighted Kohn Laplacian, which in turn is unitarily equivalent to a Schödinger operator with matrix-valued potential.
We prove new pointwise bounds for weighted Bergman kernels in $\mathbb{C}^n$, whenever a coercivity condition is satisfied by the associated weighted Kohn Laplacian on $(0,1)$-forms. Our results extend the ones obtained in $\mathbb{C}$ by Christ. Our main idea is to develop a version of Agmon theory (originally introduced to deal with Schrödinger operators) for weighted Kohn Laplacians on $(0,1)$-forms, inspired by the fact that these are unitarily equivalent to certain generalized Schrödinger operators.
We prove a family of $L^p$ uncertainty inequalities on fairly general groups and homogeneous spaces, both in the smooth and in the discrete setting. The crucial point is the proof of the $L^1$ endpoint, which is derived from a general weak isoperimetric inequality.