arXiv · cs.AI · Shaily Roy, Harshit Sharma, Daniel A. Adler, Srijan Sen, Tanzeem Choudhury, Asif Salekin · 5 小时前模型评分40
FLARE：无需已知敏感属性的可穿戴健康感知公平性框架
Ethical Fairness in Ubiquitous Health Sensing without Known Attributes
导读
研究者提出无需敏感属性的公平性框架 FLARE（Fisher-guided LAtent-subgroup learning with do-no-harm REgularization），用 Fisher 信息约束损失曲率并发现潜在表现分层，再为每层单独训练模型。
分类置信度
中 · 待人工复核
- 只有单一媒体来源，需要人工复核。
来源级别：媒体与个人
应用领域：医疗与生命科学
事件类型：研究发布
证据摘录
- “We present FLARE (Fisher-guided LAtent-subgroup learning with do-no-harm REgularization), to our knowledge the first attribute-free framework to operationalize beneficence, non-maleficence, and justice in ubiquitous sensing.”
- “Label-free routing and inference take at most 8.1 ms per sample on a smartphone and 10.5 ms on a Raspberry Pi, supporting real-time on-device deployment.”
来源：arXiv · cs.AI · arxiv.org