English(EN) One-shot federated unsupervised domain adaptation with Smoothed Knowledge Distillation and teacher refinement

新框架通过知识蒸馏简化联邦域自适应

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-23 04:00

研究人员推出了一种新颖的一次性联邦无监督域自适应框架 SKD-ETR。该方法通过减少通信轮次并提高对噪声标签的鲁棒性，解决了现有联邦学习方法中的挑战。SKD-ETR 采用平滑知识蒸馏，利用源模型的平滑伪标签在目标域上训练学生模型，同时基于指数移动平均的教师精炼将知识传回源模型。在 OfficeHome 和 DomainNet 等基准数据集上的实验表明，SKD-ETR 在提高效率和鲁棒性的同时，取得了具有竞争力的性能。 AI

影响简化了域自适应的联邦学习，有可能提高去中心化 AI 系统的效率和鲁棒性。

排序理由该集群包含一篇详细介绍域自适应新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CV 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: 4 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍域自适应新方法的论文。[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: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CV TIER_1 English(EN) · Ali Abedi, Q. M. Jonathan Wu, Ning Zhang, Farhad Pourpanah · 2026-10-23 04:00

One-shot federated unsupervised domain adaptation with Smoothed Knowledge Distillation and teacher refinement

arXiv:2610.30646v1 Announce Type: new Abstract: Federated unsupervised domain adaptation (FUDA) addresses the challenge of adapting models to an unlabeled target domain using decentralized source domains while preserving data privacy. However, existing FUDA methods typically requ…

本文如何被排名

报道来源 [1]

One-shot federated unsupervised domain adaptation with Smoothed Knowledge Distillation and teacher refinement

相关实体

相关话题