arXiv subjects

Jay Shah

Publications and source records attributed to Jay Shah.

At least 19 recordsLinked to original sources

arXiv2026

AUCp: Pseudo-AUC for Inference Model Selection with Unlabeled Validation Data in Abnormality Detection

Abnormality detection is a crucial yet challenging task in medical image analysis. Distinguishing abnormalities from normal data by learning to reconstruct normal-only data alleviates the reliance on labeled datasets. However, many studies, even if unsupervised, rely on a labeled validation set to select the best model for inference from multiple training iterations. For many diseases labeled data are unavailable and substantially time consuming to obtain. To address this, AUCp - a novel metric that supports abnormality detection for unsupervised and self-supervised methods is proposed. Instead of evaluating the realism of reconstructed images to select the best of model for inference, it focuses on actual detection performance and without requiring an annotated test set. Assuming the pseudo ground truth of all unannotated samples in the test set as abnormal/positive and using traditional AUC calculation, AUCp scores are derived. Given a large and representative training set of normal samples, we show mathematical and empirical evidence that model selection using AUCp scores improves disease detection in terms of unsupervised and self-supervised methods over conventional metrics. Using two unsupervised methods for neurologic disease detection and self-supervised methods on diverse datasets, our results demonstrate that the AUCp score effectively identifies the optimal model for inference, significantly enhancing abnormality and disease detection. The corresponding implementations are available in https://github.com/mahfuzmohammad/AUCp.

cs.CV↗

arXiv2026

FlashAttention-4: Algorithm and Kernel Pipelining Co-Design for Asymmetric Hardware Scaling

Attention, as a core layer of the ubiquitous Transformer architecture, is the bottleneck for large language models and long-context applications. While FlashAttention-3 optimized attention for Hopper GPUs through asynchronous execution and warp specialization, it primarily targets the H100 architecture. The AI industry has rapidly transitioned to deploying Blackwell-based systems such as the B200 and GB200, which exhibit fundamentally different performance characteristics due to asymmetric hardware scaling: tensor core throughput doubles while other functional units (shared memory bandwidth, exponential units) scale more slowly or remain unchanged. We develop several techniques to address these shifting bottlenecks on Blackwell GPUs: (1) redesigned pipelines that exploit fully asynchronous MMA operations and larger tile sizes, (2) software-emulated exponential and conditional softmax rescaling that reduces non-matmul operations, and (3) leveraging tensor memory and the 2-CTA MMA mode to reduce shared memory traffic and atomic adds in the backward pass. We demonstrate that our method, FlashAttention-4, achieves up to 1.3$\times$ speedup over cuDNN 9.13 and 2.7$\times$ over Triton on B200 GPUs with BF16, reaching up to 1613 TFLOPs/s (71% utilization). Beyond algorithmic innovations, we implement FlashAttention-4 entirely in CuTe-DSL embedded in Python, achieving 20-30$\times$ faster compile times compared to traditional C++ template-based approaches while maintaining full expressivity.

cs.CL↗

arXiv2026

Categorical Foundations for CuTe Layouts

NVIDIA's CUTLASS library provides a robust and expressive set of methods for describing and manipulating multi-dimensional tensor data on the GPU. These methods are conceptually grounded in the abstract notion of a CuTe layout and a rich algebra of such layouts, including operations such as composition, logical product, and logical division. In this paper, we present a categorical framework for understanding this layout algebra by focusing on a naturally occurring class of tractable layouts. To this end, we define two categories Tuple and Nest whose morphisms give rise to layouts. We define a suite of operations on morphisms in these categories and prove their compatibility with the corresponding layout operations. Moreover, we give a complete characterization of the layouts which arise from our construction. Finally, we provide a Python implementation of our categorical constructions, along with tests that demonstrate alignment with CUTLASS behavior. This implementation can be found at our git repository https://github.com/ColfaxResearch/layout-categories.

cs.PL↗

arXiv2025

DinoAtten3D: Slice-Level Attention Aggregation of DinoV2 for 3D Brain MRI Anomaly Classification

Anomaly detection and classification in medical imaging are critical for early diagnosis but remain challenging due to limited annotated data, class imbalance, and the high cost of expert labeling. Emerging vision foundation models such as DINOv2, pretrained on extensive, unlabeled datasets, offer generalized representations that can potentially alleviate these limitations. In this study, we propose an attention-based global aggregation framework tailored specifically for 3D medical image anomaly classification. Leveraging the self-supervised DINOv2 model as a pretrained feature extractor, our method processes individual 2D axial slices of brain MRIs, assigning adaptive slice-level importance weights through a soft attention mechanism. To further address data scarcity, we employ a composite loss function combining supervised contrastive learning with class-variance regularization, enhancing inter-class separability and intra-class consistency. We validate our framework on the ADNI dataset and an institutional multi-class headache cohort, demonstrating strong anomaly classification performance despite limited data availability and significant class imbalance. Our results highlight the efficacy of utilizing pretrained 2D foundation models combined with attention-based slice aggregation for robust volumetric anomaly detection in medical imaging. Our implementation is publicly available at https://github.com/Rafsani/DinoAtten3D.git.

eess.IV↗

arXiv2025

CREASE-2D Analysis of Small Angle X-ray Scattering Data from Supramolecular Dipeptide Systems

In this paper, we extend a recently developed machine-learning (ML) based CREASE-2D method to analyze the entire two-dimensional (2D) scattering pattern obtained from small angle X-ray scattering measurements of supramolecular dipeptide micellar systems. Traditional analysis of such scattering data would involve use of approximate or incorrect analytical models to fit to azimuthally-averaged 1D scattering patterns that can miss the anisotropic arrangements. Analysis of the 2D scattering profiles of such micellar solutions using CREASE-2D allows us to understand both isotropic and anisotropic structural arrangements that are present in these systems of assembled dipeptides in water and in the presence of added solvents/salts. CREASE-2D outputs distributions of relevant structural features including ones that cannot be identified with existing analytical models (e.g., assembled tubes, cross-sectional eccentricity, tortuosity, orientational order). The representative three-dimensional (3D) real-space structures for the optimized values of these structural features further facilitate visualization of the structures. Through this detailed interpretation of these 2D SAXS profiles we are able to characterize the shapes of the assembled tube structures as a function of dipeptide chemistry, solution conditions with varying salts and solvents, and relative concentrations of all components. This paper demonstrates how CREASE-2D analysis of entire SAXS profiles can provide an unprecedented level of understanding of structural arrangements which has not been possible through traditional analytical model fits to the 1D SAXS data.

cond-mat.soft↗

arXiv2024

FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision

Attention, as a core layer of the ubiquitous Transformer architecture, is the bottleneck for large language models and long-context applications. FlashAttention elaborated an approach to speed up attention on GPUs through minimizing memory reads/writes. However, it has yet to take advantage of new capabilities present in recent hardware, with FlashAttention-2 achieving only 35% utilization on the H100 GPU. We develop three main techniques to speed up attention on Hopper GPUs: exploiting asynchrony of the Tensor Cores and TMA to (1) overlap overall computation and data movement via warp-specialization and (2) interleave block-wise matmul and softmax operations, and (3) block quantization and incoherent processing that leverages hardware support for FP8 low-precision. We demonstrate that our method, FlashAttention-3, achieves speedup on H100 GPUs by 1.5-2.0$\times$ with FP16 reaching up to 740 TFLOPs/s (75% utilization), and with FP8 reaching close to 1.2 PFLOPs/s. We validate that FP8 FlashAttention-3 achieves 2.6$\times$ lower numerical error than a baseline FP8 attention.

cs.LG↗

arXiv2024

AnoFPDM: Anomaly Segmentation with Forward Process of Diffusion Models for Brain MRI

Weakly-supervised diffusion models (DMs) in anomaly segmentation, leveraging image-level labels, have attracted significant attention for their superior performance compared to unsupervised methods. It eliminates the need for pixel-level labels in training, offering a more cost-effective alternative to supervised methods. However, existing methods are not fully weakly-supervised because they heavily rely on costly pixel-level labels for hyperparameter tuning in inference. To tackle this challenge, we introduce Anomaly Segmentation with Forward Process of Diffusion Models (AnoFPDM), a fully weakly-supervised framework that operates without the need of pixel-level labels. Leveraging the unguided forward process as a reference for the guided forward process, we select hyperparameters such as the noise scale, the threshold for segmentation and the guidance strength. We aggregate anomaly maps from guided forward process, enhancing the signal strength of anomalous regions. Remarkably, our proposed method outperforms recent state-of-the-art weakly-supervised approaches, even without utilizing pixel-level labels.

cs.CV↗

arXiv2023

A Case Study in CUDA Kernel Fusion: Implementing FlashAttention-2 on NVIDIA Hopper Architecture using the CUTLASS Library

We provide an optimized implementation of the forward pass of FlashAttention-2, a popular memory-aware scaled dot-product attention algorithm, as a custom fused CUDA kernel targeting NVIDIA Hopper architecture and written using the open-source CUTLASS library. In doing so, we explain the challenges and techniques involved in fusing online-softmax with back-to-back GEMM kernels, utilizing the Hopper-specific Tensor Memory Accelerator (TMA) and Warpgroup Matrix-Multiply-Accumulate (WGMMA) instructions, defining and transforming CUTLASS Layouts and Tensors, overlapping copy and GEMM operations, and choosing optimal tile sizes for the Q, K and V attention matrices while balancing the register pressure and shared memory utilization. In head-to-head benchmarks on a single H100 PCIe GPU for some common choices of hyperparameters, we observe 20-50% higher FLOPs/s over a version of FlashAttention-2 optimized for last-generation NVIDIA Ampere architecture.

cs.LG↗

arXiv2023

Fast and Facile Synthesis Route to Epitaxial Oxide Membrane Using a Sacrificial Layer

The advancement in thin-film exfoliation for synthesizing oxide membranes has opened up new possibilities for creating artificially-assembled heterostructures with structurally and chemically incompatible materials. The sacrificial layer method is a promising approach to exfoliate as-grown films from a compatible material system, allowing their integration with dissimilar materials. Nonetheless, the conventional sacrificial layers often possess intricate stoichiometry, thereby constraining their practicality and adaptability, particularly when considering techniques like Molecular Beam Epitaxy (MBE). This is where easy-to-grow binary alkaline earth metal oxides with a rock salt crystal structure are useful. These oxides, which include (Mg, Ca, Sr, Ba)O, can be used as a sacrificial layer covering a much broader range of lattice parameters compared to conventional sacrificial layers and are easily dissolvable in deionized water. In this study, we show the epitaxial growth of single-crystalline perovskite SrTiO3 (STO) on sacrificial layers consisting of crystalline SrO, BaO, and Ba1-xCaxO films, employing a hybrid MBE method. Our results highlight the rapid (< 5 minutes) dissolution of the sacrificial layer when immersed in deionized water, facilitating the fabrication of millimeter-sized STO membranes. Using high-resolution x-ray diffraction, atomic-force microscopy, scanning transmission electron microscopy, impedance spectroscopy, and scattering-type near-field optical microscopy (SNOM), we demonstrate epitaxial STO membranes with bulk-like intrinsic dielectric properties. The employment of alkaline earth metal oxides as sacrificial layers is likely to simplify membrane synthesis, particularly with MBE, thus expanding research possibilities.

physics.app-ph↗

arXiv2023

Ordinal Classification with Distance Regularization for Robust Brain Age Prediction

Age is one of the major known risk factors for Alzheimer's Disease (AD). Detecting AD early is crucial for effective treatment and preventing irreversible brain damage. Brain age, a measure derived from brain imaging reflecting structural changes due to aging, may have the potential to identify AD onset, assess disease risk, and plan targeted interventions. Deep learning-based regression techniques to predict brain age from magnetic resonance imaging (MRI) scans have shown great accuracy recently. However, these methods are subject to an inherent regression to the mean effect, which causes a systematic bias resulting in an overestimation of brain age in young subjects and underestimation in old subjects. This weakens the reliability of predicted brain age as a valid biomarker for downstream clinical applications. Here, we reformulate the brain age prediction task from regression to classification to address the issue of systematic bias. Recognizing the importance of preserving ordinal information from ages to understand aging trajectory and monitor aging longitudinally, we propose a novel ORdinal Distance Encoded Regularization (ORDER) loss that incorporates the order of age labels, enhancing the model's ability to capture age-related patterns. Extensive experiments and ablation studies demonstrate that this framework reduces systematic bias, outperforms state-of-art methods by statistically significant margins, and can better capture subtle differences between clinical groups in an independent AD dataset. Our implementation is publicly available at https://github.com/jaygshah/Robust-Brain-Age-Prediction.

eess.IV↗

arXiv2023

Body-mounted MR-conditional Robot for Minimally Invasive Liver Intervention

MR-guided microwave ablation (MWA) has proven effective in treating hepatocellular carcinoma (HCC) with small-sized tumors, but the state-of-the-art technique suffers from sub-optimal workflow due to speed and accuracy of needle placement. This paper presents a compact body-mounted MR-conditional robot that can operate in closed-bore MR scanners for accurate needle guidance. The robotic platform consists of two stacked Cartesian XY stages, each with two degrees of freedom, that facilitate needle guidance. The robot is actuated using 3D-printed pneumatic turbines with MR-conditional bevel gear transmission systems. Pneumatic valves and control mechatronics are located inside the MRI control room and are connected to the robot with pneumatic transmission lines and optical fibers. Free space experiments indicated robot-assisted needle insertion error of 2.6$\pm$1.3 mm at an insertion depth of 80 mm. The MR-guided phantom studies were conducted to verify the MR-conditionality and targeting performance of the robot. Future work will focus on the system optimization and validations in animal trials.

cs.RO↗

arXiv2023

Symmetry breaking and ascending in the magnetic kagome metal FeGe

Spontaneous symmetry breaking-the phenomenon where an infinitesimal perturbation can cause the system to break the underlying symmetry-is a cornerstone concept in the understanding of interacting solid-state systems. In a typical series of temperature-driven phase transitions, higher temperature phases are more symmetric due to the stabilizing effect of entropy that becomes dominant as the temperature is increased. However, the opposite is rare but possible when there are multiple degrees of freedom in the system. Here, we present such an example of a symmetry-ascending phenomenon in a magnetic kagome metal FeGe by utilizing neutron Larmor diffraction and Raman spectroscopy. In the paramagnetic state at 460K, we confirm that the crystal structure is indeed hexagonal kagome lattice. On cooling to TN, the crystal structure changes from hexagonal to monoclinic with in-plane lattice distortions on the order of 10^(-4) and the associated splitting of the double degenerate phonon mode of the pristine kagome lattice. Upon further cooling to TCDW, the kagome lattice shows a small negative thermal expansion, and the crystal structure becomes more symmetric gradually upon further cooling. Increasing the crystalline symmetry upon cooling is unusual, it originates from an extremely weak structural instability that coexists and competes with the CDW and magnetic orders. These observations are against the expectations for a simple model with a single order parameter, hence can only be explained by a Landau free energy expansion that takes into account multiple lattice, charge, and spin degrees of freedom. Thus, the determination of the crystalline lattice symmetry as well as the unusual spin-lattice coupling is a first step towards understanding the rich electronic and magnetic properties of the system and sheds new light on intertwined orders where the lattice degree of freedom is no longer dominant.

cond-mat.str-el↗

arXiv2023

Brainomaly: Unsupervised Neurologic Disease Detection Utilizing Unannotated T1-weighted Brain MR Images

Harnessing the power of deep neural networks in the medical imaging domain is challenging due to the difficulties in acquiring large annotated datasets, especially for rare diseases, which involve high costs, time, and effort for annotation. Unsupervised disease detection methods, such as anomaly detection, can significantly reduce human effort in these scenarios. While anomaly detection typically focuses on learning from images of healthy subjects only, real-world situations often present unannotated datasets with a mixture of healthy and diseased subjects. Recent studies have demonstrated that utilizing such unannotated images can improve unsupervised disease and anomaly detection. However, these methods do not utilize knowledge specific to registered neuroimages, resulting in a subpar performance in neurologic disease detection. To address this limitation, we propose Brainomaly, a GAN-based image-to-image translation method specifically designed for neurologic disease detection. Brainomaly not only offers tailored image-to-image translation suitable for neuroimages but also leverages unannotated mixed images to achieve superior neurologic disease detection. Additionally, we address the issue of model selection for inference without annotated samples by proposing a pseudo-AUC metric, further enhancing Brainomaly's detection performance. Extensive experiments and ablation studies demonstrate that Brainomaly outperforms existing state-of-the-art unsupervised disease and anomaly detection methods by significant margins in Alzheimer's disease detection using a publicly available dataset and headache detection using an institutional dataset. The code is available from https://github.com/mahfuzmohammad/Brainomaly.

eess.IV↗

arXiv2022

HealthyGAN: Learning from Unannotated Medical Images to Detect Anomalies Associated with Human Disease

Automated anomaly detection from medical images, such as MRIs and X-rays, can significantly reduce human effort in disease diagnosis. Owing to the complexity of modeling anomalies and the high cost of manual annotation by domain experts (e.g., radiologists), a typical technique in the current medical imaging literature has focused on deriving diagnostic models from healthy subjects only, assuming the model will detect the images from patients as outliers. However, in many real-world scenarios, unannotated datasets with a mix of both healthy and diseased individuals are abundant. Therefore, this paper poses the research question of how to improve unsupervised anomaly detection by utilizing (1) an unannotated set of mixed images, in addition to (2) the set of healthy images as being used in the literature. To answer the question, we propose HealthyGAN, a novel one-directional image-to-image translation method, which learns to translate the images from the mixed dataset to only healthy images. Being one-directional, HealthyGAN relaxes the requirement of cycle consistency of existing unpaired image-to-image translation methods, which is unattainable with mixed unannotated data. Once the translation is learned, we generate a difference map for any given image by subtracting its translated output. Regions of significant responses in the difference map correspond to potential anomalies (if any). Our HealthyGAN outperforms the conventional state-of-the-art methods by significant margins on two publicly available datasets: COVID-19 and NIH ChestX-ray14, and one institutional dataset collected from Mayo Clinic. The implementation is publicly available at https://github.com/mahfuzmohammad/HealthyGAN.

eess.IV↗

arXiv2022

Parametrized and equivariant higher algebra

We develop the rudiments of a theory of parametrized $\infty$-operads, including parametrized generalizations of monoidal envelopes, Day convolution, operadic left Kan extensions, results on limits and colimits of algebras, and the symmetric monoidal Yoneda embedding.

math.AT↗

arXiv2021

On the equivalence of two theories of real cyclotomic spectra

We give a new formula for real topological cyclic homology that refines the fiber sequence formula discovered by Nikolaus and Scholze for topological cyclic homology to one involving genuine $C_2$-spectra. To accomplish this, we give a new definition of the $\infty$-category of real cyclotomic spectra that replaces the usage of genuinely equivariant dihedral spectra with the parametrized Tate construction. We then define an $\infty$-categorical version of H{\o}genhaven's O(2)-orthogonal cyclotomic spectra, construct a forgetful functor relating the two theories, and show that this functor restricts to an equivalence between full subcategories of appropriately bounded-below objects. As an application, we compute the real topological cyclic homology of perfect $\mathbb{F}_p$-algebras for all primes $p$.

math.AT↗

arXiv2021

Process Design and Economics of Production of p-Aminophenol

Para-Aminophenol is one of the key chemicals required for the synthesis of Paracetamol, an analgesic and antipyretic drug. Data shows a large fraction of India's demand for Para-Aminophenol being met through imports from China. The uncertainty in the India-China relations would affect the supply and price of this "Key Starting Material." This report is a detailed business plan for setting up a plant and producing Para-Aminophenol in India at a competitive price. The plant is simulated in AspenPlus V8 and different Material Balances and Energy Balances calculations are carried out. The plant produces 22.7 kmols Para-Aminophenol per hour with a purity of 99.9%. Along with the simulation, economic analysis is carried out for this plant to determine the financial parameters like Payback Period and Return on Investment.

econ.GN↗

arXiv2021

On the parametrized Tate construction

We introduce and study a genuine equivariant refinement of the Tate construction associated to an extension $\widehat{G}$ of a finite group $G$ by a compact Lie group $K$, which we call the parametrized Tate construction $(-)^{t_G K}$. Our main theorem establishes the coincidence of three conceptually distinct approaches to its construction when $K$ is also finite: one via recollement theory for the $K$-free $\widehat{G}$-family, another via parametrized ambidexterity for $G$-local systems, and the last via parametrized assembly maps. We also show that $(-)^{t_G K}$ uniquely admits the structure of a lax $G$-symmetric monoidal functor, thereby refining a theorem of Nikolaus and Scholze. Along the way, we apply a theorem of the second author to reprove a result of Ayala--Mazel-Gee--Rozenblyum on reconstructing a genuine $G$-spectrum from its geometric fixed points; our method of proof further yields a formula for the geometric fixed points of an $\mathcal{F}$-complete $G$-spectrum for any $G$-family $\mathcal{F}$.

math.AT↗