Google Research has introduced SensorFM, a large sensor foundation model trained on over a trillion minutes of data from five million individuals. This model learns a general representation of human physiology from unlabeled wearable data, enabling it to transfer across various health prediction tasks and serve as a grounding tool for a Personal Health Agent. SensorFM utilizes self-supervised learning with an Adaptive and Inherited Masking framework to effectively handle missing or fragmented data common in wearable devices. AI
IMPACT Could enable more personalized and preventive healthcare by leveraging large-scale wearable sensor data.
RANK_REASON Research paper detailing a new foundation model for wearable health data. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Google AI / Research →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
