Search

We present an end-to-end automated workflow that uses large-scale remote compute resources and an embedded GPU platform at the edge to enable AI/ML-accelerated real-time analysis of data collected for x-ray ptychography. Ptychography is a lensless method that is being used to image samples through a simultaneous numerical inversion of a large number of diffraction patterns from adjacent overlapping scan positions. This acquisition method can enable nanoscale imaging with x-rays and electrons, but this often requires very large experimental datasets and commensurately high turnaround times, which can limit experimental capabilities such as real-time experimental steering and low-latency monitoring. In this work, we introduce a software system that can automate ptychography data analysis tasks. We accelerate the data analysis pipeline by using a modified version of PtychoNN -- an ML-based approach to solve phase retrieval problem that shows two orders of magnitude speedup compared to traditional iterative methods. Further, our system coordinates and overlaps different data analysis tasks to minimize synchronization overhead between different stages of the workflow. We evaluate our workflow system with real-world experimental workloads from the 26ID beamline at Advanced Photon Source and ThetaGPU cluster at Argonne Leadership Computing Resources., Comment: 7 pages, 1 figure, to be published in High Performance Computing for Imaging Conference, Electronic Imaging (HPCI 2023)

Coherent microscopy techniques provide an unparalleled multi-scale view of materials across scientific and technological fields, from structural materials to quantum devices, from integrated circuits to biological cells. Driven by the construction of brighter sources and high-rate detectors, coherent X-ray microscopy methods like ptychography are poised to revolutionize nanoscale materials characterization. However, associated significant increases in data and compute needs mean that conventional approaches no longer suffice for recovering sample images in real-time from high-speed coherent imaging experiments. Here, we demonstrate a workflow that leverages artificial intelligence at the edge and high-performance computing to enable real-time inversion on X-ray ptychography data streamed directly from a detector at up to 2 kHz. The proposed AI-enabled workflow eliminates the sampling constraints imposed by traditional ptychography, allowing low dose imaging using orders of magnitude less data than required by traditional methods.

We show that the generalized Reynolds number (in fluid dynamics) introduced by Ladyzhenskaya is closely related to the rotation of the positive spectral subspace of the Stokes block-operator in the underlying Hilbert space. We also explicitly evaluate the bottom of the negative spectrum of the Stokes operator and prove a sharp inequality relating the distance from the bottom of its spectrum to the origin and the length of the first positive gap., 19 pages, 2 figures

We consider a Schr��dinger operator with bounded, measurable potential in multidimensional Euclidean space. We prove for every $L^2$-eigenfunction a quantitative equidistribution estimate. It compares the total $L^2$-norm with the $L^2$-norm over an equidistributed collection of balls. Our estimate is explicit with respect to the radius of the balls, norm of the potential and the energy of the eigenfunction. Similar estimates also hold for Weyl sequences and for linear combinations of eigenfunctions, as long as the associated eigenvalues are sufficiently close., Short note of 4 pages. The first version of the paper had a mistake in the M-scaling. This is now corrected

We review recent results on localization for discrete alloy-type models based on the multiscale analysis and the fractional moment method, respectively. The discrete alloy-type model is a family of Schr��dinger operators $H_��= - ��+ V_��$ on $\ell^2 (\ZZ^d)$ where $��$ is the discrete Laplacian and $V_��$ the multiplication by the function $V_��(x) = \sum_{k \in \ZZ^d} ��_k u(x-k)$. Here $��_k$, $k \in \ZZ^d$, are i.i.d. random variables and $u \in \ell^1 (\ZZ^d ; \RR)$ is a so-called single-site potential. Since $u$ may change sign, certain properties of $H_��$ depend in a non-monotone way on the random parameters $��_k$. This requires new methods at certain stages of the localization proof., Proceedings of the Spectral Days 2010, Pontificia Universidad Cat\'olica de Chile, Santiago

We study spectral properties of Schr��dinger operators on $\RR^d$. The electromagnetic potential is assumed to be determined locally by a colouring of the lattice points in $\ZZ^d$, with the property that frequencies of finite patterns are well defined. We prove that the integrated density of states (spectral distribution function) is approximated by its finite volume analogues, i.e.the normalised eigenvalue counting functions. The convergence holds in the space $L^p(I)$ where $I$ is any finite energy interval and $1\leq p< \infty$ is arbitrary., 15 pages; v2 has minor fixes

A version of the Davis-Kahan Tan $2��$ theorem [SIAM J. Numer. Anal. \textbf{7} (1970), 1 -- 46] for not necessarily semibounded linear operators defined by quadratic forms is proven. This theorem generalizes a recent result by Motovilov and Selin [Integr. Equat. Oper. Theory \textbf{56} (2006), 511 -- 542].

We consider a finite dimensional damped second order system and obtain spectral inclusion theorems for the related quadratic eigenvalue problem. The inclusion sets are the 'quasi Cassini ovals' which may greatly outperform standard Gershgorin circles. As the unperturbed system we take a modally damped part of the system; this includes the known proportionally damped models, but may give much sharper estimates. These inclusions are then applied to derive some easily calculable sufficient conditions for the overdampedness of a given damped system.

We consider an alloy type potential on an infinite metric graph. We assume a covering condition on the single site potentials. For random Schr��dingers operator associated with the alloy type potential restricted to finite volume subgraphs we prove a Wegner estimate which reproduces the modulus of continuity of the single site distribution measure. The Wegner constant is independent of the energy., 8 pages

We consider ergodic random magnetic Schr��dinger operators on the metric graph $\mathbb{Z}^d$ with random potentials and random boundary conditions taking values in a finite set. We show that normalized finite volume eigenvalue counting functions converge to a limit uniformly in the energy variable. This limit, the integrated density of states, can be expressed by a closed Shubin-Pastur type trace formula. It supports the spectrum and its points of discontinuity are characterized by existence of compactly supported eigenfunctions. Among other examples we discuss percolation models., 17 pages; typos removed, references updated, definition of subgraph densities explained

The concept of Lorentz scalar quark confinement has a long history and is still widely used despite its well-known theoretical faults. We point out here that the predictions of scalar confinement also conflict directly with experiment. We investigate the dependence of heavy-light meson mass differences on the mass of the light quark. In particular, we examine the strange and non-strange D mesons. We find that the predictions of scalar confinement are in considerable conflict with measured values., Comment: REVTeX4, 7 pages, 4 EPS figures

Clean predictions are presented for all the spin-averaged heavy-light meson spectroscopies. A new symmetry is identified wherein the energy eigenstates have a universal dependence on both the light and heavy quark masses. This universality is used in an efficient analysis of these mesons within the QCD string/flux tube picture. Unique predictions for all the D, D_s, B, and B_s type mesons in terms of just four measured quantities., Comment: REVTeX4, 6 pages, 9 eps figures

The CDF and D0 experiments probe the high-energy frontier and as they do so have accumulated hundreds of Terabytes of data on the way to petabytes of data over the next two years. The experiments have made a commitment to use the developing Grid based on the SAM system to handle these data. The D0 SAM has been extended for use in CDF as common patterns of design emerged to meet the similar requirements of these experiments. The process by which the merger was achieved is explained with particular emphasis on lessons learned concerning the database design patterns plus realization of the use cases., Talk from the 2003 Computing in High Energy and Nuclear Physics (CHEP03), La Jolla, Ca, USA, March 2003, 4 pages, pdf format, TUAT004

We consider the Riemannian universal covering of a compact manifold $M = X / ��$ and assume that $��$ is amenable. We show for an ergodic random family of Schr��dinger operators on $X$ the existence of a (non-random) integrated density of states., LaTeX 2e, amsart, 17 pages; appeared in a somewhat different form in Geometriae Dedicata, 91 (1): 117-135, (2002)

We perform an analytic semi-classical quantization of the straight QCD string with one end fixed and a massless quark on the other, in the limits of orbital and radial dominant motion. We compare our results to the exact numerical semi-classical quantization. We observe that the numerical semi-classical quantization agrees well with our exact numerical canonical quantization., Comment: RevTeX, 10 pages, 9 figures

We consider the effects of going beyond the approximation of a straight string in mesons by using a flexible flux tube model wherein a Nambu-Goto string bends in response to quark accelerations. The curved string is dynamically identical to the straight string even for ultra-relativistic mesons except for a small additional radial momentum. We numerically solve the curved string model in the case where both ends have equal mass quarks and also the case where one end is fixed. No approximation of non-relativistic motion is made. We note some small but interesting difference from the straight string., Comment: 12 pages REVTeX, 10 PostScript Figures

Recent lattice results for the energy of gluonic excitations as a function of quark separation are shown to correspond to transverse relativistic flux tube vibration modes. For large quark separations all states appear to degenerate into a few categories which are predicted uniquely, given the ground state., Comment: 10 pages, 3 figures, LaTeX, references added and minor corrections

We calculate the strong radiative corrections to the process p pbar -> W(->e nu) g^*(->b bbar). At the Tevatron this process is the largest background to the associated Higgs boson production p pbar -> W(-> e nu) H(-> b bbar). The calculation is based on the subtraction procedure, and the corrections are found to be significant., 8 pages, latex, 4 postscript figures included

We consider several issues related to the multidimensional integration using a network of heterogeneous computers. Based on these considerations, we develop a new general purpose scheme which can significantly reduce the time needed for evaluation of integrals with CPU intensive integrands. This scheme is a parallel version of the well-known adaptive Monte Carlo method (the VEGAS algorithm), and is incorporated into a new integration package which uses the standard set of message-passing routines in the PVM software system., Comment: 19 pages, latex, 5 postscript figures included

Starting from Buchm\"uller's observation that a chromoelectric flux tube meson will exhibit only the Thomas type spin-orbit interaction, we show that a model built upon the related assumption that a quark feels only a constant radial chromoelectric field in its rest frame implies a complete relativistic effective Hamiltonian that can be written explicitly in terms of quark canonical variables. The model yields linear Regge trajectories and exhibits some similarities to scalar confinement, but with the advantage of being more closely linked to QCD., Comment: 12 pages, LaTeX 2.09, uses RevTeX and AMSfonts. To be published in Phys Rev D

Hybrid charmonium with mass ~4 GeV could be produced via a c c.bar color-octet component in b->c c.bar s. These states could be narrow and could have a significant branching ratio to light hadrons, perhaps enhanced by glueballs. Decays to gluonic hadrons could make a sizable contribution to B->no charm decays. Experimental signatures and search strategies are discussed., 12 pages, latex

We apply HQET to semi-leptonic $B$ and $B_{s}$ meson decays into the observed charmed $P$ wave states. In order to examine the sensitivity of the results to the choice of a specific model, we perform all calculations using several different meson models, and find that uncertainty introduced by the choice of a particular model is about 30\%. Specifically, assuming $\tau_{B}=1.50ps$ and $V_{cb}= 0.040$, we obtain branching ratios of $(0.27\pm 0.08)\%$ and $(0.45\pm 0.14)\%$ for $B\rar D_{1}l\bar{\nu}_{l}$ and $B\rar D_{2}^{*}l\bar{\nu}_{l}$ decays, respectively., Comment: Latex (uses epsf macro), 36 pages of text, 11 postscript figures included

We reexamine contributions of higher $K$-resonances to the radiative rare decays $b\rar s\gamma$ in the limit where both $b$- and $s$-quark are considered heavy. Using the non-relativistic quark model, and the form factor definitions consistent with the HQET covariant trace formalism, we find significant disagreement with previous work which also used heavy quark symmetry, and excellent agreement with experimental results. In particular, the two largest fractions of the inclusive $b\rar s\gamma$ branching ratio are found to be $(16.8\pm 6.4)\%$ for $B\rar K^{*}(892)\gamma$ and $(6.2\pm 2.9)\%$ for $B\rar K_{2}^{*}(1430)\gamma$ decays. We also compare the contribution from the radiative decays into the eight $K$-meson states to the inclusive experimental $b\rar s\gamma$ mass distribution., Comment: Latex (uses epsf macro), 20 pages of text, 1 postscript figure included

We adapt a general method to solve both the full and reduced Salpeter equations and systematically explore the conditions under which these two equations give equivalent results in meson dynamics. The effects of constituent mass, angular momentum state, type of interaction, and the nature of confinement are all considered in an effort to clearly delineate the range of validity of the reduced Salpeter approximations. We find that for $J\not{\hspace*{-1.0mm}=}0$ the solutions are strikingly similar for all constituent masses. For zero angular momentum states the full and reduced Salpeter equations give different results for small quark mass especially with a large additive constant coordinate space potential. We also show that $\frac{1}{m}$ corrections to heavy-light energy levels can be accurately computed with the reduced equation., Comment: Latex (uses epsf macro), 24 pages of text, 12 postscript figures included. Slightly revised version, to appear in Phys. Rev. D

We present a systematic algebraic and numerical investigation of the instantaneous Bethe-Salpeter equation. Emphasis is placed on confining interaction kernels of the Lorentz scalar, time component vector, and full vector types. We explore stability of the solutions and Regge behavior for each of these interactions, and conclude that only time component vector confinement leads to normal Regge structure and stable solutions., Comment: Latex (uses epsf macro), 26 pages of text, 12 postscript figures included.

The Relativistic Flux Tube model is used to calculate of the Isgur-Wise functions describing the exclusive semileptonic decays of $\bar{B}$ and $\bar{B}_{s}$ mesons. The light quark mass dependence is investigated and the predicted universal function agrees well with the results of lattice simulations, and with the experimental data. Recent experimental measurements of the $\bar{B}\ra D^{*}l\bar{\nu}_{l}$ decay distribution yield the CKM element $V_{bc}=0.035\pm 0.001$. The IW function slope and second derivative at the zero recoil point are predicted to be $\xi'(1) = -0.93\pm 0.04$ and $\xi''(1)= 1.7\pm 0.1$ for a range of light quark massses. The importance of including higher derivatives in analyses of experiment is emphasized., Comment: Latex (uses epsf macro), 14 pages of text, 6 postscript figures included. MAD/PH/851