arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2410.13638cs.LGcs.AIcs.HC
收藏

可穿戴基础模型的规模化扩展

Scaling Wearable Foundation Models

Google Research(谷歌研究院);Google DeepMind(谷歌DeepMind)
Girish Narayanswamy, Xin Liu, Kumar Ayush, Yuzhe Yang, Xuhai Xu, Shun Liao, Jake Garrison, Shyam Tailor, Jake Sunshine, Yun Liu, Tim Althoff, Shrikanth Narayana… 展开作者

Girish Narayanswamy, Xin Liu, Kumar Ayush, Yuzhe Yang, Xuhai Xu, Shun Liao, Jake Garrison, Shyam Tailor, Jake Sunshine, Yun Liu, Tim Althoff, Shrikanth Narayanan, Pushmeet Kohli, Jiening Zhan, Mark Malhotra, Shwetak Patel, Samy Abdel-Ghaffar, Daniel McDuff

AI总结:

本文基于超过165,000人、4000万小时的多模态可穿戴传感器数据构建了基础模型LSM,并系统研究了其在计算、数据和模型规模上的扩展规律,展示了该模型在信号插补、外推及下游活动识别任务中实现样本高效学习的能力。

摘要:

得益于多种健康追踪功能,可穿戴传感器已变得无处不在。由此产生的来自日常生活的连续纵向测量数据生成了海量数据;然而,从这些观测中提取科学且可操作的见解并非易事。受生成建模经验性成功的启发——其中大型神经网络从海量文本、图像、视频或音频数据中学习强大的表示——我们研究了传感器基础模型在计算、数据和模型规模上的扩展特性。我们使用来自超过165,000人的多达4000万小时的原位心率、心率变异性、皮肤电活动、加速度计、皮肤温度和高度计每分钟数据,创建了LSM,这是一个多模态基础模型,构建于迄今为止最大的可穿戴信号数据集之上,并拥有最广泛的传感器模态范围。我们的结果确立了LSM在插补、插值和外推等任务上的扩展规律,这些任务既跨时间也跨传感器模态。此外,我们强调了LSM如何为运动和活动识别等任务实现样本高效的下游学习。

英文摘要:

Wearable sensors have become ubiquitous thanks to a variety of health tracking features. The resulting continuous and longitudinal measurements from everyday life generate large volumes of data; however, making sense of these observations for scientific and actionable insights is non-trivial. Inspired by the empirical success of generative modeling, where large neural networks learn powerful representations from vast amounts of text, image, video, or audio data, we investigate the scaling properties of sensor foundation models across compute, data, and model size. Using a dataset of up to 40 million hours of in-situ heart rate, heart rate variability, electrodermal activity, accelerometer, skin temperature, and altimeter per-minute data from over 165,000 people, we create LSM, a multimodal foundation model built on the largest wearable-signals dataset with the most extensive range of sensor modalities to date. Our results establish the scaling laws of LSM for tasks such as imputation, interpolation and extrapolation, both across time and sensor modalities. Moreover, we highlight how LSM enables sample-efficient downstream learning for tasks like exercise and activity recognition.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑