English(EN) State-of-art minibatches via novel DPP kernels: discretization, wavelets, and rough objectives

新的DPP核利用小波改进机器学习小批量

作者 PulseAugur 编辑部 · [2 个来源] · 2026-05-13 07:54

研究人员开发了使用小波的新型行列式点过程（DPP），以改进机器学习任务的小批量生成。这些新颖的DPP提供了可证明的更优准确性保证，并提供了一种将连续DPP转换为适合子采样的离散核的通用方法。该方法提高了方差缩减和计算效率，将基于DPP的方法的适用性扩展到正则性较低的目标函数。 AI

影响引入了一种新颖的方法，用于在机器学习中生成更有效、更准确的小批量，有可能提高训练性能并降低计算成本。

排序理由该集群包含一篇详细介绍机器学习新方法的学术论文。

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv stat.ML TIER_1 English(EN) · Hoang-Son Tran, Pranav Gupta, R\'emi Bardenet, Subhroshekhar Ghosh · 2026-05-14 04:00

State-of-art minibatches via novel DPP kernels: discretization, wavelets, and rough objectives

arXiv:2605.13127v1 Announce Type: new Abstract: Determinantal point processes (DPPs) have emerged as a kernelized alternative to vanilla independent sampling for generating efficient minibatches, coresets and other parsimonious representations of large-scale datasets. While theor…
arXiv stat.ML TIER_1 English(EN) · Subhroshekhar Ghosh · 2026-05-13 07:54

State-of-art minibatches via novel DPP kernels: discretization, wavelets, and rough objectives

Determinantal point processes (DPPs) have emerged as a kernelized alternative to vanilla independent sampling for generating efficient minibatches, coresets and other parsimonious representations of large-scale datasets. While theoretical foundations and promising empirical perfo…