IIFL: Implicit Interactive Fleet Learning from Heterogeneous Human Supervisors
This addresses the challenge of handling diverse human demonstrations in robotic fleet learning, representing an incremental improvement over existing methods.
The paper tackles the problem of distribution shift in imitation learning for robots by proposing Implicit Interactive Fleet Learning (IIFL), which learns from multiple heterogeneous human supervisors using energy-based models and a novel uncertainty quantification method, achieving a 2.8x higher success rate in simulation and 4.5x higher return on human effort in a physical task compared to baselines.
Imitation learning has been applied to a range of robotic tasks, but can struggle when robots encounter edge cases that are not represented in the training data (i.e., distribution shift). Interactive fleet learning (IFL) mitigates distribution shift by allowing robots to access remote human supervisors during task execution and learn from them over time, but different supervisors may demonstrate the task in different ways. Recent work proposes Implicit Behavior Cloning (IBC), which is able to represent multimodal demonstrations using energy-based models (EBMs). In this work, we propose Implicit Interactive Fleet Learning (IIFL), an algorithm that builds on IBC for interactive imitation learning from multiple heterogeneous human supervisors. A key insight in IIFL is a novel approach for uncertainty quantification in EBMs using Jeffreys divergence. While IIFL is more computationally expensive than explicit methods, results suggest that IIFL achieves a 2.8x higher success rate in simulation experiments and a 4.5x higher return on human effort in a physical block pushing task over (Explicit) IFL, IBC, and other baselines.