arXiv subjects

Aashwin Mishra

Publications and source records attributed to Aashwin Mishra.

15 recordsLinked to original sources

arXiv2026

Towards single-shot coherent imaging via overlap-free ptychography

Ptychographic imaging at synchrotron and XFEL sources requires dense overlapping scans, limiting throughput and increasing dose. Extending coherent diffractive imaging to overlap-free operation on extended samples remains an open problem. Here, we extend PtychoPINN (O. Hoidn \emph{et al.}, \emph{Scientific Reports} \textbf{13}, 22789, 2023) to deliver \emph{overlap-free, single-shot} reconstructions in a Fresnel coherent diffraction imaging (CDI) geometry while also accelerating conventional multi-shot ptychography. The framework couples a differentiable forward model of coherent scattering with a Poisson photon-counting likelihood; real-space overlap enters as a tunable parameter via coordinate-based grouping rather than a hard requirement. On synthetic benchmarks, reconstructions remain accurate at low counts ($\sim\!10^4$ photons/frame), and overlap-free single-shot reconstruction with an experimental probe reaches amplitude structural similarity (SSIM) 0.904, compared with 0.968 for overlap-constrained reconstruction. Against a data-saturated supervised model with the same backbone (16,384 training images), PtychoPINN achieves higher SSIM with only 1,024 images and generalizes to unseen illumination profiles. Per-graphics processing unit (GPU) throughput is approximately $40\times$ that of least-squares maximum-likelihood (LSQ-ML) reconstruction at matched $128\times128$ resolution. These results, validated on experimental data from the Advanced Photon Source and the Linac Coherent Light Source, unify single-exposure Fresnel CDI and overlapped ptychography within one framework, supporting dose-efficient, high-throughput imaging at modern light sources.

physics.optics↗

arXiv2026

Domain Knowledge Guided Bayesian Optimization For Autonomous Alignment Of Complex Scientific Instruments

Bayesian Optimization (BO) is a powerful tool for optimizing complex non-linear systems. However, its performance degrades in high-dimensional problems with tightly coupled parameters and highly asymmetric objective landscapes, where rewards are sparse. In such needle-in-a-haystack scenarios, even advanced methods like trust-region BO (TurBO) often lead to unsatisfactory results. We propose a domain knowledge guided Bayesian Optimization approach, which leverages physical insight to fundamentally simplify the search problem by transforming coordinates to decouple input features and align the active subspaces with the primary search axes. We demonstrate this approach's efficacy on a challenging 12-dimensional, 6-crystal Split-and-Delay optical system, where conventional approaches, including standard BO, TuRBO and multi-objective BO, consistently led to unsatisfactory results. When combined with an reverse annealing exploration strategy, this approach reliably converges to the global optimum. The coordinate transformation itself is the key to this success, significantly accelerating the search by aligning input co-ordinate axes with the problem's active subspaces. As increasingly complex scientific instruments, from large telescopes to new spectrometers at X-ray Free Electron Lasers are deployed, the demand for robust high-dimensional optimization grows. Our results demonstrate a generalizable paradigm: leveraging physical insight to transform high-dimensional, coupled optimization problems into simpler representations can enable rapid and robust automated tuning for consistent high performance while still retaining current optimization algorithms.

cs.LG↗

arXiv2025

Deep Learning Enabled Nanoscale X-ray Photoemission Electron Microscopy (nanoXPEEM)

Understanding and manipulating two-dimensional materials for real-world applications remains challenging due to a lack of effective and high-throughput characterization techniques. Soft X-ray time-of-flight photoemission electron microscopy (XPEEM) provides element- and depth-sensitive information of materials and buried interfaces. However, chromatic and spherical aberrations cannot be corrected with electron-lens combinations. These aberrations, combined with astigmatism and space-charge effects, significantly degrade the spatial and energy resolutions. To overcome this limitation, we outline a spatial-attention based deep learning approach to automatically correct for these effects and attain nanometer resolution over the entire field-of-view (FoV). The combination of this corrective algorithm with XPEEM, termed as nanoXPEEM, establishes a new record of 48-nm spatial resolution with a 232-micrometer diameter FoV in the soft x-ray regime (700-1000 eV). nanoXPEEM provides unique spatial mapping of the element-specificity, depth-sensitivity, and local structure on the nanoscale. It can bridge the current gap to achieve angstrom (atomic) scale resolution.

cond-mat.mtrl-sci↗

arXiv2025

Data Driven Drift Correction For Complex Optical Systems

To exploit the thousand-fold increase in spectral brightness of modern light sources, increasingly intricate experiments are being conducted that demand extremely precise beam trajectory. Maintaining the optimal trajectory over several hours of an experiment with the needed precision necessitates active drift control. Here, we outline Time-Varying Bayesian Optimization (TVBO) as a data driven approach for robust drift correction, and illustrate its application for a split and delay optical system composed of six crystals and twelve input dimensions. Using numerical simulations, we exhibit the application of TVBO for linear drift, non-smooth temporal drift as well as constrained TVBO for multi-objective control settings, representing real-life operating conditions. This approach can be easily adapted to other X-ray beam conditioning and guidance systems, including multi-crystal monochromators and grazing-incidence mirrors, to maintain sub-micron and nanoradian beam stability over the course of an experiment spanning several hours.

physics.optics↗

arXiv2025

A Start To End Machine Learning Approach To Maximize Scientific Throughput From The LCLS-II-HE

With the increasing brightness of Light sources, including the Diffraction-Limited brightness upgrade of APS and the high-repetition-rate upgrade of LCLS, the proposed experiments therein are becoming increasingly complex. For instance, experiments at LCLS-II-HE will require the X-ray beam to be within a fraction of a micron in diameter, with pointing stability of a few nanoradians, at the end of a kilometer-long electron accelerator, a hundred-meter-long undulator section, and tens of meters long X-ray optics. This enhancement of brightness will increase the data production rate to rival the largest data generators in the world. Without real-time active feedback control and an optimized pipeline to transform measurements to scientific information and insights, researchers will drown in a deluge of mostly useless data, and fail to extract the highly sophisticated insights that the recent brightness upgrades promise. In this article, we outline the strategy we are developing at SLAC to implement Machine Learning driven optimization, automation and real-time knowledge extraction from the electron-injector at the start of the electron accelerator, to the multidimensional X-ray optical systems, and till the experimental endstations and the high readout rate, multi-megapixel detectors at LCLS to deliver the design performance to the users. This is illustrated via examples from Accelerator, Optics and End User applications.

physics.ins-det↗

arXiv2025

Uncertainty Propagation within Chained Models for Machine Learning Reconstruction of Neutrino-LAr Interactions

Sequential or chained models are increasingly prevalent in machine learning for scientific applications, due to their flexibility and ease of development. Chained models are particularly useful when a task is separable into distinct steps with a hierarchy of meaningful intermediate representations. In reliability-critical tasks, it is important to quantify the confidence of model inferences. However, chained models pose an additional challenge for uncertainty quantification, especially when input uncertainties need to be propagated. In such cases, a fully uncertainty-aware chain of models is required, where each step accepts a probability distribution over the input space, and produces a probability distribution over the output space. In this work, we present a case study for adapting a single model within an existing chain, designed for reconstruction within neutrino-Argon interactions, developed for neutrino oscillation experiments such as MicroBooNE, ICARUS, and the future DUNE experiment. We test the performance of an input uncertainty-enabled model against an uncertainty-blinded model using a method for generating synthetic noise. By comparing these two, we assess the increase in inference quality achieved by exposing models to upstream uncertainty estimates.

physics.data-an↗

arXiv2024

Generalizing Stochastic Smoothing for Differentiation and Gradient Estimation

We deal with the problem of gradient estimation for stochastic differentiable relaxations of algorithms, operators, simulators, and other non-differentiable functions. Stochastic smoothing conventionally perturbs the input of a non-differentiable function with a differentiable density distribution with full support, smoothing it and enabling gradient estimation. Our theory starts at first principles to derive stochastic smoothing with reduced assumptions, without requiring a differentiable density nor full support, and we present a general framework for relaxation and gradient estimation of non-differentiable black-box functions $f:\mathbb{R}^n\to\mathbb{R}^m$. We develop variance reduction for gradient estimation from 3 orthogonal perspectives. Empirically, we benchmark 6 distributions and up to 24 variance reduction strategies for differentiable sorting and ranking, differentiable shortest-paths on graphs, differentiable rendering for pose estimation, as well as differentiable cryo-ET simulations.

cs.LG↗

arXiv2024

Uncertainty Quantification via Stable Distribution Propagation

We propose a new approach for propagating stable probability distributions through neural networks. Our method is based on local linearization, which we show to be an optimal approximation in terms of total variation distance for the ReLU non-linearity. This allows propagating Gaussian and Cauchy input uncertainties through neural networks to quantify their output uncertainties. To demonstrate the utility of propagating distributions, we apply the proposed method to predicting calibrated confidence intervals and selective prediction on out-of-distribution data. The results demonstrate a broad applicability of propagating distributions and show the advantages of our method over other approaches such as moment matching.

cs.LG↗

arXiv2023

Probabilistic Mixture Model-Based Spectral Unmixing

Identifying pure components in mixtures is a common yet challenging problem. The associated unmixing process requires the pure components, also known as endmembers, to be sufficiently spectrally distinct. Even with this requirement met, extracting the endmembers from a single mixture is impossible; an ensemble of mixtures with sufficient diversity is needed. Several spectral unmixing approaches have been proposed, many of which are connected to hyperspectral imaging. However, most of them assume highly diverse collections of mixtures and extremely low-loss spectroscopic measurements. Additionally, non-Bayesian frameworks do not incorporate the uncertainty inherent in unmixing. We propose a probabilistic inference approach that explicitly incorporates noise and uncertainty, enabling us to unmix endmembers in collections of mixtures with limited diversity. We use a Bayesian mixture model to jointly extract endmember spectra and mixing parameters while explicitly modeling observation noise and the resulting inference uncertainties. We obtain approximate distributions over endmember coordinates for each set of observed spectra while remaining robust to inference biases from the lack of pure observations and presence of non-isotropic Gaussian noise. Access to reliable uncertainties on the unmixing solutions would enable robust solutions as well as informed decision making.

physics.data-an↗

arXiv2023

Deep Neural Network Uncertainty Quantification for LArTPC Reconstruction

We evaluate uncertainty quantification (UQ) methods for deep learning applied to liquid argon time projection chamber (LArTPC) physics analysis tasks. As deep learning applications enter widespread usage among physics data analysis, neural networks with reliable estimates of prediction uncertainty and robust performance against overconfidence and out-of-distribution (OOD) samples are critical for their full deployment in analyzing experimental data. While numerous UQ methods have been tested on simple datasets, performance evaluations for more complex tasks and datasets are scarce. We assess the application of selected deep learning UQ methods on the task of particle classification using the PiLArNet [1] monte carlo 3D LArTPC point cloud dataset. We observe that UQ methods not only allow for better rejection of prediction mistakes and OOD detection, but also generally achieve higher overall accuracy across different task settings. We assess the precision of uncertainty quantification using different evaluation metrics, such as distributional separation of prediction entropy across correctly and incorrectly identified samples, receiver operating characteristic curves (ROCs), and expected calibration error from observed empirical accuracy. We conclude that ensembling methods can obtain well calibrated classification probabilities and generally perform better than other existing methods in deep learning UQ literature.

hep-ex↗

arXiv2023

Machine Learning Based Alignment For LCLS-II-HE Optics

The hard X-ray instruments at the Linac Coherent Light Source are in the design phase for upgrades that will take full advantage of the high repetition rates that will become available with LCLS-II-HE. The current X-ray Correlation Spectroscopy instrument will be converted to the Dynamic X-ray Scattering instrument, and will feature a meV-scale high-resolution monochromator at its front end with unprecedented coherent flux. With the new capability come many engineering and design challenges, not least of which is the sensitivity to long-term drift of the optics. With this in mind, we have estimated the system tolerance to angular drift and vibration for all the relevant optics ($\sim$10 components) in terms of how the central energy out of the monochromator will be affected to inform the mechanical design. Additionally, we have started planning for methods to correct for such drifts using available (both invasive and non-invasive) X-ray beam diagnostics. In simulations, we have demonstrated the ability of trained Machine Learning models to correct misalignments to maintain the desired central energy and optical axis within the necessary tolerances. Additionally, we exhibit the use of Bayesian Optimization to minimize the impact of thermal deformations of crystals as well as beam alignment from scratch. The initial results are very promising and efforts to further extend this work are ongoing.

physics.acc-ph↗

arXiv2022

Testing the data framework for an AI algorithm in preparation for high data rate X-ray facilities

The advent of next-generation X-ray free electron lasers will be capable of delivering X-rays at a repetition rate approaching 1 MHz continuously. This will require the development of data systems to handle experiments at these type of facilities, especially for high throughput applications, such as femtosecond X-ray crystallography and X-ray photon fluctuation spectroscopy. Here, we demonstrate a framework which captures single shot X-ray data at the LCLS and implements a machine-learning algorithm to automatically extract the contrast parameter from the collected data. We measure the time required to return the results and assess the feasibility of using this framework at high data volume. We use this experiment to determine the feasibility of solutions for `live' data analysis at the MHz repetition rate.

physics.data-an↗

arXiv2022

A machine learning photon detection algorithm for coherent X-ray ultrafast fluctuation analysis

X-ray free electron laser (XFEL) experiments have brought unique capabilities and opened new directions in research, such as creating new states of matter or directly measuring atomic motion. One such area is the ability to use finely spaced sets of coherent x-ray pulses to be compared after scattering from a dynamic system at different times. This enables the study of fluctuations in many-body quantum systems at the level of the ultrafast pulse durations, but this method has been limited to a select number of examples and required complex and advanced analytical tools. By applying a new methodology to this problem, we have made qualitative advances in three separate areas that will likely also find application to new fields. As compared to the `droplet-type' models which typically are used to estimate the photon distributions on pixelated detectors to obtain the coherent X-ray speckle patterns, our algorithm pipeline achieves an order of magnitude speedup on CPU hardware and two orders of magnitude improvement on GPU hardware. We also find that it retains accuracy in low-contrast conditions, which is the typical regime for many experiments in structural dynamics. Finally, it can predict photon distributions in high average-intensity applications, a regime which up until now, has not been accessible. Our AI-assisted algorithm will enable a wider adoption of x-ray coherence spectroscopies, by both automating previously challenging analyses and enabling new experiments that were not otherwise feasible without the developments described in this work.

physics.data-an↗

arXiv2021

Improving Surrogate Model Accuracy for the LCLS-II Injector Frontend Using Convolutional Neural Networks and Transfer Learning

Machine learning models of accelerator systems (`surrogate models') are able to provide fast, accurate predictions of accelerator physics phenomena. However, approaches to date typically do not include measured input diagnostics, such as the initial beam distributions, which are critical for accurately representing the beam evolution through the system. In addition, these inputs often vary over time, and models that can account for these changing conditions are needed. Simulations are in some cases needed to provide sufficient training data. These typically represent the designed machine before construction; however, the behavior of the installed components may be quite different due to changes over time or static differences that were not modeled. Therefore, surrogate models that can leverage both simulation and measured data successfully are needed. We introduce an approach based on convolutional neural networks that uses the drive laser distribution and scalar settings as inputs for the Linac Coherent Light Source II injector frontend. The model is able to predict scalar beam parameters and the transverse beam distribution downstream, taking into account the impact of time-varying non-uniformities in the initial transverse laser distribution. We also introduce and evaluate a transfer learning procedure for adapting the surrogate model from the simulation domain to the measurement domain, to account for differences between the two. Applying this approach to our test case results in a model that can predict test sample outputs within a mean absolute percent error of 7.6%. This is a substantial improvement over the model trained only on simulations, which has an error of 112.7% when applied to measured data. While we focus on the LCLS-II Injector frontend, these approaches for improving ML-based online modeling of injector systems could be easily adapted to other accelerator facilities.

physics.acc-ph↗

arXiv2016

Manufactured Turbulence with Langevin equations

By definition, Manufactured turbulence(MT) is purported to mimic physical turbulence rather than model it. The MT equations are constrained to be simple to solve and provide an inexpensive surrogate to Navier-Stokes based Direct Numerical Simulations (DNS) for use in engineering applications or theoretical analyses. In this article, we investigate one approach in which the linear inviscid aspects of MT are derived from a linear approximation of the Navier-Stokes equations while the non-linear and viscous physics are approximated via stochastic modeling. The ensuing Langevin MT equations are used to compute planar, quadratic turbulent flows. While much work needs to be done, the preliminary results appear promising.

physics.flu-dyn↗