发表机构
New York University; Columbia University; Oregon Health & Science University（纽约大学; 哥伦比亚大学; 俄勒冈健康与科学大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文发现纵向对比学习中记录增长会扩大不变性、抑制变化信息，提出显式对比同人其他观测可逆转损失，在2.7年数据和199名参与者中验证。
AI 中文摘要
纵向数据之所以有价值，是因为人会变化。然而，用于从这些数据中学习的目标函数可能会无意中抹去这种变化。在个体级对比学习中，来自同一个人的观测被视为正样本对；随着记录的增长，这些正样本可能跨越越来越遥远——且越来越不同——的行为状态。因此，更多的历史数据不仅可能带来更多的数据，还可能带来更广泛的不变性。我们证明这种区分是根本性的。我们将“记录跨度”（学习者能看到多少历史）与“监督跨度”（正样本对监督在该历史中延伸多远）分离开来。在跨越长达2.7年的居家传感记录中，更广泛的监督系统地抑制了可恢复的变化状态信息，即使可用历史保持不变。在最宽的跨度下，从未经训练的编码器中可恢复的信息中，只有不到10%得以保留。然而，仅仅保持正样本局部化并不足够：随着记录增长，即使从未配对过的遥远状态也变得越来越相似。显式地对比来自同一个人的其他观测，可以在不缩短记录的情况下逆转这种损失，揭示了纵向规模扩大不变性的第二条途径。最后，我们在199名GLOBEM参与者中前瞻性地重现了监督跨度效应。因此，纵向规模呈现出一个选择：更多的历史并不必然意味着更广泛的不变性。通过控制随着记录增长而保持不变的内容，我们可以保留最初使纵向数据有价值的变化。
英文摘要
Longitudinal data are valuable because people change. Yet the objectives used to learn from these data can inadvertently erase that change. In person-level contrastive learning, observations from the same person are treated as positives; as records grow, those positives can span increasingly distant---and increasingly different---behavioral states. More history can therefore produce not only more data, but broader invariance. We show that this distinction is fundamental. We separate \emph{record span}, how much history the learner sees, from \emph{supervision span}, how far across that history positive-pair supervision reaches. Across in-home sensing records spanning up to 2.7 years, broader supervision systematically suppresses recoverable changing-state information, even when the available history is held fixed. At the broadest span, less than 10\% of the information recoverable from an untrained encoder remains. Yet keeping positives local is not sufficient: as records grow, even distant states that are never paired become increasingly similar. Explicitly contrasting other observations from the same person reverses this loss without shortening the record, revealing a second route by which longitudinal scale can broaden invariance. Finally, we prospectively reproduce the supervision-span effect in 199 GLOBEM participants. Longitudinal scale therefore presents a choice: more history need not mean more invariance. By controlling what is held invariant as records grow, we can preserve the change that made the longitudinal data valuable in the first place.
Comments53 pages, 6 figures, 54 tables
