用于预测性人类建模的解耦技能表示

中文摘要 AI 辅助

理解人类技能对于与人类协作、指导或提供帮助的AI系统十分重要。与依赖单一观测的典型潜变量估计问题不同，技能是一种持久、可组合且基于行为的构造，必须通过随时间变化的模式进行推断。我们引入了Skill Abstraction with Interpretable Latents（SAIL，具有可解释潜变量的技能抽象），这是一种将人类技能建模为从自然行为中推断出的可解释多维构造的方法。我们的方法生成的技能嵌入对瞬时性能波动具有鲁棒性，并学习人类子技能的可迁移表示。此外，SAIL支持跨多种域内上下文泛化的技能感知行为预测。我们用一个持久的技能嵌入来表示每个个体，该嵌入控制专家与新手基础之间的融合，并使用反事实子技能交换进行解耦训练。这种设计鼓励表示既对性能变化具有鲁棒性，又具有可解释性的结构化。我们在赛车和棒球任务中证明，SAIL实现了强大的预测性能，且相较于评估的基线方法，在基于行为的解耦方面始终表现更优，同时还提升了下游AI指导性能。

英文摘要

Understanding human skill is important for AI systems that collaborate with, coach, or assist people. Unlike typical latent variable estimation problems which rely on single observations, skill is a persistent, compositional, and behaviorally grounded construct that must be inferred from patterns over time. We introduce Skill Abstraction with Interpretable Latents (SAIL), a method for modeling human skill as an interpretable, multi-dimensional construct inferred from naturalistic behavior. Our approach produces a skill embedding that is robust to transient performance fluctuations and learns a transferable representation of human subskills. Furthermore, SAIL supports skill-informed behavior prediction that generalizes across a variety of in-domain contexts. We represent each individual with a persistent skill embedding that controls a blend between expert and novice bases and is trained using counterfactual subskill swaps for disentanglement. This design encourages representations that are both robust to performance variation and structured for interpretability. We demonstrate across racing and baseball that SAIL achieves strong predictive performance and consistently improves behaviorally grounded disentanglement over the evaluated baselines, while also improving downstream AI coaching performance.

发表机构

• Toyota Research Institute（丰田研究所）

机构由 AI 辅助整理，请以论文原文为准。