AI Coaching for Accelerating Human Skill Development with Reinforcement Learning
For researchers and practitioners in human-AI interaction and skill training, this work provides a principled approach to designing AI coaches that avoid over-reliance and promote long-term skill development.
This paper introduces an AI coaching framework that uses reinforcement learning to accelerate human motor-skill development by strategically scaffolding and stepping back based on learner capability. In a user study on drone racing (N=33), the method significantly outperformed state-of-the-art AI coaching baselines in improving human learning outcomes.
AI copilots can substantially boost human performance through shared control, but excessive assistance can induce over-reliance and skill atrophy. This paper studies how an embodied AI agent can act as a coach that accelerates human motor-skill development. We argue that effective coaching requires strategic scaffolding and stepping back that are aligned with the learner's capability, allowing productive failures that drive learning. We formalize the interactive AI coaching process as a non-cooperative dynamic game in which the learner optimizes task performance while the coach targets the learner's independent competence. Building on this formalism, we develop a reinforcement learning framework combining adaptive shared control with probabilistic models of the coach's causal influence on skill evolution, enabling tractable training of coaching policies. A comprehensive user study (N=33) on first-person-view drone racing shows significant gains in human learning outcomes over state-of-the-art AI coaching baselines.