English(EN) PACE: Propagation-Aware Collaborative Correction for One-Shot Personalized Federated Graph Learning

新的PACE方法增强了个性化联邦图学习

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-07 04:00

研究人员开发了PACE，一种用于个性化联邦图学习的新颖方法，该方法解决了客户端异构性问题。PACE将协同知识视为对本地模型的紧凑校正，而不是替代，允许接收者保留其完整的本地模型。该方法使用秩-r更新载体和对角线草图进行通信，并采用凸负对数似然校准（CNLL）来选择本地和外部对数之间的系数。PACE在多个数据集上显示出准确性和加权F1的改进，尤其是在外部校正被认为不合适时保留本地预测。 AI

影响为改进基于图的人工智能系统中的联邦学习引入了一项新技术。

排序理由详细介绍图学习新方法的学术论文。[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: 32 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍图学习新方法的学术论文。[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) · Ruizhe Huang, Chengran Li, Xiaochuan Shi · 2026-09-07 04:00

PACE：面向一次性个性化联邦图学习的传播感知协同校正

arXiv:2609.04832v1 Announce Type: new Abstract: Client heterogeneity creates both an opportunity and a risk in personalized federated graph learning. Knowledge held by other subgraphs may complement a receiver's Local model, but an incompatible transfer can override reliable pred…

本文如何被排名

报道来源 [1]

PACE：面向一次性个性化联邦图学习的传播感知协同校正

相关实体

相关话题