Easy-IIL: Reducing Human Operational Burden in Interactive Imitation Learning via Assistant Experts

2026-03-13

AI Summary

Interactive imitation learning (IIL) typically imposes a heavy operational burden by requiring extensive offline demonstrations and frequent online human interventions. This work proposes Easy-IIL, a novel framework that, for the first time, employs a model-based imitation learning agent as an auxiliary expert to autonomously collect the majority of training data, thereby reducing human involvement to only an initial demonstration and targeted interventions during critical failure states. By integrating an end-to-end policy, a human-in-the-loop intervention mechanism, and techniques for maintaining data quality, Easy-IIL substantially alleviates human effort in both simulated and real-world environments while achieving performance on par with state-of-the-art IIL baselines. User studies further confirm a significant reduction in perceived workload, demonstrating the framework’s practicality and usability.

Technology Category

• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Intelligent Robots: Human-Robot Interaction
• Humans and AI: Human-in-the-loop Machine Learning

Application Category

• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Interactive Imitation Learning (IIL) typically relies on extensive human involvement for both offline demonstration and online interaction. Prior work primarily focuses on reducing human effort in passive monitoring rather than active operation. Interestingly, structured model-based imitation approaches achieve comparable performance with significantly fewer demonstrations than end-to-end imitation learning policies in the low-data regime. However, these methods are typically surpassed by end-to-end policies as the data increases. Leveraging this insight, we propose Easy-IIL, a framework that utilizes off-the-shelf model-based imitation methods as an assistant expert to replace active human operation for the majority of data collection. The human expert only provides a single demonstration to initialize the assistant expert and intervenes in critical states where the task is approaching failure. Furthermore, Easy-IIL can maintain IIL performance by preserving both offline and online data quality. Extensive simulation and real-world experiments demonstrate that Easy-IIL significantly reduces human operational burden while maintaining performance comparable to mainstream IIL baselines. User studies further confirm that Easy-IIL reduces subjective workload on the human expert. Project page: https://sites.google.com/view/easy-iil

Problem

• Interactive Imitation Learning
• Human Operational Burden
• Assistant Experts
• Demonstration Efficiency
• Human-in-the-loop

Innovation

• Interactive Imitation Learning
• Assistant Expert
• Model-based Imitation
• Human Operational Burden Reduction
• Critical-state Intervention