Flash-SD-KDE: Accelerating SD-KDE with Tensor Cores

• URL: http://arxiv.org/abs/2602.10378v1
• Date: Tue, 10 Feb 2026 23:56:03 GMT
• Title: Flash-SD-KDE: Accelerating SD-KDE with Tensor Cores
• Abstract summary: We show that by re-ordering the SD-KDE to expose matrix-multiplication structure, Cores can be used to accelerate the GPU implementation.<n>On a 1M-sample 16-dimensional task evaluated on 131k queries, Flash-SD-KDE completes in $ s on a single GPU, making score-debiased density estimation practical at previously infeasible scales.
• Score: 2.421459418045937
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Score-debiased kernel density estimation (SD-KDE) achieves improved asymptotic convergence rates over classical KDE, but its use of an empirical score has made it significantly slower in practice. We show that by re-ordering the SD-KDE computation to expose matrix-multiplication structure, Tensor Cores can be used to accelerate the GPU implementation. On a 32k-sample 16-dimensional problem, our approach runs up to $47\times$ faster than a strong SD-KDE GPU baseline and $3{,}300\times$ faster than scikit-learn's KDE. On a larger 1M-sample 16-dimensional task evaluated on 131k queries, Flash-SD-KDE completes in $2.3$ s on a single GPU, making score-debiased density estimation practical at previously infeasible scales.

Related papers

• FlashPDE: A Drop-In Fused Triton Operator Library for Neural PDE Solvers [6.9997407868865364]
  FlashPDE is a drop-in fused operator library for grid-based machine learning.<n>It replaces fragmented PyTorch finite-difference execution with differentiable Triton kernels.<n>It reduces peak memory usage by up to 37.0x compared with coordinate-based automatic differentiation.
  arXiv  Detail & Related papers  (2026-07-20T14:54:03Z)
• DepthART: Scaling Foundation Monocular Depth to Tiny Models [51.10174763031444]
  DepthART is a compact MDE model for on-device deployment across diverse scenes.<n>We show how DepthART consistently surpasses previous tiny baselines in both zero-shot generalization and metric accuracy.<n>We also provide a scalable model family, with DepthART-S reaching 347/245 FPS (strict FP32) on an GTX A6000 at $2242/4482$, 102 FPS (TF32) on a Orin NX 8GB, and over 15 FPS (FP32) on a Jetson Nano 4GB.
  arXiv  Detail & Related papers  (2026-07-19T06:54:52Z)
• SD-KDE: Score-Debiased Kernel Density Estimation [0.0]
  We propose a novel method for density estimation that leverages an estimated score function to debias kernel density estimation.<n>Our experiments on synthetic tasks in 1D, 2D and on MNIST, demonstrate that our proposed SD-KDE method significantly reduces the mean integrated squared error.
  arXiv  Detail & Related papers  (2025-04-27T02:51:30Z)
• Transolver++: An Accurate Neural Solver for PDEs on Million-Scale Geometries [67.63077028746191]
  Transolver++ is a highly parallel and efficient neural solver that can solve PDEs on million-scale geometries.<n>Transolver++ increases the single- GPU input capacity to million-scale points for the first time.<n>It achieves over 20% performance gain in million-scale high-fidelity industrial simulations.
  arXiv  Detail & Related papers  (2025-02-04T15:33:50Z)
• DPOT: Auto-Regressive Denoising Operator Transformer for Large-Scale PDE Pre-Training [87.90342423839876]
  We present a new auto-regressive denoising pre-training strategy, which allows for more stable and efficient pre-training on PDE data. We train our PDE foundation model with up to 0.5B parameters on 10+ PDE datasets with more than 100k trajectories.
  arXiv  Detail & Related papers  (2024-03-06T08:38:34Z)
• Elucidating the solution space of extended reverse-time SDE for diffusion models [23.637881476921596]
  We formulate the sampling process as an Extended Reverse-Time SDE (ER SDE)<n>We offer exact solutions and approximate solutions for SDE and VE SDE, respectively.<n>We devise efficient high-quality samplers, namely ER-SDE-rs.
  arXiv  Detail & Related papers  (2023-09-12T12:27:17Z)
• Fast Private Kernel Density Estimation via Locality Sensitive Quantization [10.227538355037554]
  We study efficient mechanisms for differentially private kernel density estimation (DP-KDE) We show how the kernel can privately be approximated in time linear in $d$, making it feasible for high-dimensional data.
  arXiv  Detail & Related papers  (2023-07-04T18:48:04Z)
• HDCC: A Hyperdimensional Computing compiler for classification on embedded systems and high-performance computing [58.720142291102135]
  This work introduces the name compiler, the first open-source compiler that translates high-level descriptions of HDC classification methods into optimized C code. name is designed like a modern compiler, featuring an intuitive and descriptive input language, an intermediate representation (IR), and a retargetable backend. To substantiate these claims, we conducted experiments with HDCC on several of the most popular datasets in the HDC literature.
  arXiv  Detail & Related papers  (2023-04-24T19:16:03Z)
• KDEformer: Accelerating Transformers via Kernel Density Estimation [30.860357184928407]
  In this paper, we introduce a new approach to exact computation of Dot-product attention mechanism. We show that our proposed approach outperforms other attention approximations in terms of accuracy, memory, and runtime. For image classification with T2T-ViT, we show over $18times$ speedup while the accuracy drop is less than $0.5%$.
  arXiv  Detail & Related papers  (2023-02-05T18:23:49Z)
• Dynamic Maintenance of Kernel Density Estimation Data Structure: From Practice to Theory [16.460386321121945]
  Kernel density estimation (KDE) stands out as a challenging task in machine learning. Recently, there has been a growing trend of using data structures for efficient KDE. In this work, we focus on the dynamic maintenance of KDE data structures with robustness to adversarial queries.
  arXiv  Detail & Related papers  (2022-08-08T04:40:20Z)
• DEANN: Speeding up Kernel-Density Estimation using Approximate Nearest Neighbor Search [8.25574589820305]
  We present an algorithm called Density Estimation from Approximate Nearest Neighbors (DEANN) We apply ANN algorithms as a black box subroutine to compute an unbiased Density Estimation (KDE) We show that our implementation outperforms state of the art implementations in all high dimensional datasets we considered.
  arXiv  Detail & Related papers  (2021-07-06T17:11:28Z)
• GBHT: Gradient Boosting Histogram Transform for Density Estimation [73.94900378709023]
  We propose a density estimation algorithm called textitGradient Boosting Histogram Transform (GBHT) We make the first attempt to theoretically explain why boosting can enhance the performance of its base learners for density estimation problems.
  arXiv  Detail & Related papers  (2021-06-10T13:40:28Z)
• Efficient and Generic 1D Dilated Convolution Layer for Deep Learning [52.899995651639436]
  We introduce our efficient implementation of a generic 1D convolution layer covering a wide range of parameters. It is optimized for x86 CPU architectures, in particular, for architectures containing Intel AVX-512 and AVX-512 BFloat16 instructions. We demonstrate the performance of our optimized 1D convolution layer by utilizing it in the end-to-end neural network training with real genomics datasets.
  arXiv  Detail & Related papers  (2021-04-16T09:54:30Z)
• Kernel Stein Generative Modeling [68.03537693810972]
  Gradient Langevin Dynamics (SGLD) demonstrates impressive results with energy-based models on high-dimensional and complex data distributions. Stein Variational Gradient Descent (SVGD) is a deterministic sampling algorithm that iteratively transports a set of particles to approximate a given distribution. We propose noise conditional kernel SVGD (NCK-SVGD), that works in tandem with the recently introduced Noise Conditional Score Network estimator.
  arXiv  Detail & Related papers  (2020-07-06T21:26:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.