EpiCoach：用 LLM 智能体把个体运动员当作"N-of-1"科学实验对象的可审计生理建模
The Athlete as a Scientist of One: Agentic Bayesian Experimental Design for Individualized Physiology from Consumer Wearables
AI 导读
研究者提出 EpiCoach，一个基于 LLM 智能体的贝叶斯实验设计系统，将一名山地自行车运动员作为可审计、可证伪的 N-of-1 科学对象。系统由心率、GPS 和气压海拔数据推断出机械式数字孪生（滞后心率响应模型、带学习时间常数的体能-疲劳模型、潜在下坡技术状态），智能体以最大化信息增益的约束实验开出处方。
正文 · 原文
Abstract:Large language model (LLM) agents for personal health read wearable data and recommend, but do not learn about the individual in a way that can be audited or falsified. We present EpiCoach, an agentic system that treats one athlete, a cross-country mountain biker, as the subject of accountable N-of-1 science. Claims about the athlete live in an evidence-gated hypothesis registry: each must issue a probabilistic prediction about a future session before it occurs, and its belief moves only through computed evidence from a logged statistical fit. Predictions come from a mechanistic digital twin inferred from heart rate, GPS and barometric altitude: a lagged heart-rate response model, a fitness-fatigue model with learned time constants, and a latent technical-descending skill state, measured on discovered trail segments. The agent prescribes the next training week as a constrained experiment maximizing information gain subject to load and safety constraints, around two instrument rides. We evaluate the system two ways over eighteen months of data. A rolling-origin replay benchmark shows the twin beats standard forecasters on heart rate but not segment time, and the agentic loop matches a static model. A live prospective phase runs the weekly cycle: in its first three weeks the ledger sealed 20 predictions, scored 14, the descent forecast calibrated after one correction, the climb forecast pessimistic as the athlete improved, and the safety officer vetoed the agent twice. Every correction is preserved in the registry, a template for auditable, falsifiable personalization in digital health. We release the registry, benchmark and protocol.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.40370 [cs.RO] |
| (or arXiv:2610.40370v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.40370 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Behnam Moradi [view email]
[v1]
Thu, 29 Oct 2026 16:20:24 UTC (1,357 KB)
来源：arXiv cs.AI · arxiv.org