English(EN) Greedy Orthogonal Polynomial Manifolds with Lipschitz Control for Dimensionality Reduction and Reduced Modeling

新方法Greedy-OPM通过Lipschitz控制增强降维性能

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-28 04:00

研究人员推出了一种用于降维的新方法Greedy-OPM，该方法利用多项式流形。该方法使用从训练数据派生的标准正交特征来表示多项式项。Greedy-OPM同时选择编码器模式和多项式特征，旨在最小化重建误差，同时遵守解码器的Lipschitz常数界限。研究表明，这种Lipschitz约束增强了泛化能力并保持了高重建精度。 AI

排序理由该集群包含一篇关于降维新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 15 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇关于降维新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, other; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Tomoki Koike, Benjamin Peherstorfer, Elizabeth Qian · 2026-10-28 04:00

用于降维和降阶建模的具有 Lipschitz 控制的贪婪正交多项式流形

arXiv:2610.33784v1 Announce Type: new Abstract: Polynomial manifolds provide nonlinear low-dimensional representations through polynomial decoders. We propose to represent the polynomial terms of the decoder using features that are orthonormal on the training data. This empirical…

本文如何被排名

报道来源 [1]

用于降维和降阶建模的具有 Lipschitz 控制的贪婪正交多项式流形

相关实体

相关话题