研究人员开发了一种新的框架，用于在可穿戴设备上个性化人类活动识别（HAR）模型。这种无梯度的方法可以重新利用现有的HAR分类器，用最少的校准数据适应新用户，即使这些数据是未标记的或不可用的。该方法在准确性方面显示出显著的改进，监督自适应将性能提高了高达33.44个百分点，无监督自适应提高了高达32.13个百分点。 AI
影响 通过有限的用户数据，实现更准确、更高效的可穿戴AI模型在设备端的个性化。
排序理由 该集群包含一篇详细介绍机器学习个性化新方法的论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2606.04798v1 Announce Type: new Abstract: Sensor-based Human Activity Recognition (HAR) models often degrade on unseen users due to domain shifts caused by individual movement patterns and sensor placement. Practical wearable HAR systems therefore require personalization me…
Sensor-based Human Activity Recognition (HAR) models often degrade on unseen users due to domain shifts caused by individual movement patterns and sensor placement. Practical wearable HAR systems therefore require personalization methods that are lightweight, applicable whether c…