Aashwin Mishra
YOU?
Author Swipe
View article: A Fragile Number Sense: Probing the Elemental Limits of Numerical Reasoning in LLMs
A Fragile Number Sense: Probing the Elemental Limits of Numerical Reasoning in LLMs Open
Large Language Models (LLMs) have demonstrated remarkable emergent capabilities, yet the robustness of their numerical reasoning remains an open question. While standard benchmarks evaluate LLM reasoning on complex problem sets using aggre…
View article: Data Driven Drift Correction For Complex Optical Systems
Data Driven Drift Correction For Complex Optical Systems Open
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 hou…
View article: Reversing the Lens: Using Explainable AI to Understand Human Expertise
Reversing the Lens: Using Explainable AI to Understand Human Expertise Open
Both humans and machine learning models learn from experience, particularly in safety- and reliability-critical domains. While psychology seeks to understand human cognition, the field of Explainable AI (XAI) develops methods to interpret …
View article: Network Models of Expertise in the Complex Task of Operating Particle Accelerators
Network Models of Expertise in the Complex Task of Operating Particle Accelerators Open
We implement a network-based approach to study expertise in a complex real-world task: operating particle accelerators. Most real-world tasks we learn and perform (e.g., driving cars, operating complex machines, solving mathematical proble…
View article: Generalizing Stochastic Smoothing for Differentiation and Gradient Estimation
Generalizing Stochastic Smoothing for Differentiation and Gradient Estimation Open
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-differe…
View article: Active Learning for Rapid Targeted Synthesis of Compositionally Complex Alloys
Active Learning for Rapid Targeted Synthesis of Compositionally Complex Alloys Open
The next generation of advanced materials is tending toward increasingly complex compositions. Synthesizing precise composition is time-consuming and becomes exponentially demanding with increasing compositional complexity. An experienced …
View article: Probabilistic Mixture Model-Based Spectral Unmixing
Probabilistic Mixture Model-Based Spectral Unmixing Open
Spectral unmixing attempts to decompose a spectral ensemble into the constituent pure spectral signatures (called endmembers) along with the proportion of each endmember. This is essential for techniques like hyperspectral imaging (HSI) us…
View article: Active Learning for Rapid Targeted Synthesis of Compositionally Complex Alloys
Active Learning for Rapid Targeted Synthesis of Compositionally Complex Alloys Open
The next generation of advanced materials is tending toward increasingly complex compositions. Synthesizing precise composition is time-consuming and becomes exponentially demanding with increasing compositional complexity. An experienced …
View article: Uncertainty Quantification via Stable Distribution Propagation
Uncertainty Quantification via Stable Distribution Propagation Open
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 …
View article: Physics constrained unsupervised deep learning for rapid, high resolution scanning coherent diffraction reconstruction
Physics constrained unsupervised deep learning for rapid, high resolution scanning coherent diffraction reconstruction Open
By circumventing the resolution limitations of optics, coherent diffractive imaging (CDI) and ptychography are making their way into scientific fields ranging from X-ray imaging to astronomy. Yet, the need for time consuming iterative phas…
View article: Probabilistic Mixture Model-Based Spectral Unmixing
Probabilistic Mixture Model-Based Spectral Unmixing Open
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, ex…
View article: Machine Learning Based Alignment For LCLS-II-HE Optics
Machine Learning Based Alignment For LCLS-II-HE Optics Open
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 Spectro…
View article: Physics Constrained Unsupervised Deep Learning for Rapid, High Resolution Scanning Coherent Diffraction Reconstruction
Physics Constrained Unsupervised Deep Learning for Rapid, High Resolution Scanning Coherent Diffraction Reconstruction Open
By circumventing the resolution limitations of optics, coherent diffractive imaging (CDI) and ptychography are making their way into scientific fields ranging from X-ray imaging to astronomy. Yet, the need for time consuming iterative phas…
View article: Deep Neural Network Uncertainty Quantification for LArTPC Reconstruction
Deep Neural Network Uncertainty Quantification for LArTPC Reconstruction Open
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…
View article: Low-energy Electron-track Imaging for a Liquid Argon Time-projection-chamber Telescope Concept Using Probabilistic Deep Learning
Low-energy Electron-track Imaging for a Liquid Argon Time-projection-chamber Telescope Concept Using Probabilistic Deep Learning Open
The GammaTPC is an MeV-scale single-phase liquid argon time-projection-chamber gamma-ray telescope concept with a novel dual-scale pixel-based charge-readout system. It promises to enable a significant improvement in sensitivity to MeV-sca…
View article: Gamma Ray Source Localization for Time Projection Chamber Telescopes Using Convolutional Neural Networks
Gamma Ray Source Localization for Time Projection Chamber Telescopes Using Convolutional Neural Networks Open
Diverse phenomena such as positron annihilation in the Milky Way, merging binary neutron stars, and dark matter can be better understood by studying their gamma ray emission. Despite their importance, MeV gamma rays have been poorly explor…
View article: Testing the data framework for an AI algorithm in preparation for high data rate X-ray facilities
Testing the data framework for an AI algorithm in preparation for high data rate X-ray facilities Open
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 fac…
View article: A machine learning photon detection algorithm for coherent x-ray ultrafast fluctuation analysis
A machine learning photon detection algorithm for coherent x-ray ultrafast fluctuation analysis Open
X-ray free electron laser 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…
View article: A machine learning photon detection algorithm for coherent X-ray ultrafast fluctuation analysis
A machine learning photon detection algorithm for coherent X-ray ultrafast fluctuation analysis Open
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 spac…
View article: Simulated X-ray Photon Fluctuation Spectroscopy Dataset
Simulated X-ray Photon Fluctuation Spectroscopy Dataset Open
Dataset for simulated X-ray Photon Fluctuation Spectroscopy (XPFS) detector images and photon maps. XPFS is a X-ray speckle imaging technique used at SLAC National Accelerator Laboratory to study ultrafast materials dynamics.
View article: Simulated X-ray Photon Fluctuation Spectroscopy Dataset
Simulated X-ray Photon Fluctuation Spectroscopy Dataset Open
Dataset for simulated X-ray Photon Fluctuation Spectroscopy (XPFS) detector images and photon maps. XPFS is a X-ray speckle imaging technique used at SLAC National Accelerator Laboratory to study ultrafast materials dynamics.
View article: Combustion machine learning: Principles, progress and prospects
Combustion machine learning: Principles, progress and prospects Open
View article: Uncertainty quantification for deep learning in particle accelerator applications
Uncertainty quantification for deep learning in particle accelerator applications Open
With the advent of increased computational resources and improved algorithms, machine learning-based models are being increasingly applied to complex problems in particle accelerators. However, such data-driven models may provide overly co…
View article: Interpretable data-driven methods for subgrid-scale closure in LES for transcritical LOX/GCH4 combustion
Interpretable data-driven methods for subgrid-scale closure in LES for transcritical LOX/GCH4 combustion Open
View article: Improving surrogate model accuracy for the LCLS-II injector frontend using convolutional neural networks and transfer learning
Improving surrogate model accuracy for the LCLS-II injector frontend using convolutional neural networks and transfer learning Open
Machine learning (ML) 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…
View article: Estimating RANS model uncertainty using machine learning
Estimating RANS model uncertainty using machine learning Open
In this work we present a machine-learning strategy developed to estimate the uncertainty introduced by a turbulence model for the prediction of a turbulent separated flows. The approach is based on the introduction of eigenvalue perturbat…
View article: Improving Surrogate Model Accuracy for the LCLS-II Injector Frontend Using Convolutional Neural Networks and Transfer Learning
Improving Surrogate Model Accuracy for the LCLS-II Injector Frontend Using Convolutional Neural Networks and Transfer Learning Open
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 t…
View article: Data-assisted combustion simulations with dynamic submodel assignment using random forests
Data-assisted combustion simulations with dynamic submodel assignment using random forests Open
View article: Measurement-Based Surrogate Model of the SLAC LCLS-II Injector
Measurement-Based Surrogate Model of the SLAC LCLS-II Injector Open
There is significant effort within particle accelerator physics to use machine learning methods to improve modeling of accelerator components. Such models can be made realistic and representative of machine components by training them with…
View article: An uncertainty estimation module for turbulence model predictions in SU2
An uncertainty estimation module for turbulence model predictions in SU2 Open
With the advent of improved computational resources, aerospace design has testing-based process to a simulation-driven procedure, wherein uncertainties in design and operating conditions are explicitly accounted for in the design under unc…