AdaDemo: Data-Efficient Demonstration Expansion for Generalist Robotic Agent
This work addresses the data efficiency challenge in imitation learning for robotics, offering an incremental improvement in demonstration collection methods.
The paper tackles the problem of scaling up demonstrations for generalist robotic agents by introducing AdaDemo, a framework that actively expands demonstration datasets to improve multi-task policy learning, achieving progressive performance improvements across 22 tasks in two robotic manipulation benchmarks.
Encouraged by the remarkable achievements of language and vision foundation models, developing generalist robotic agents through imitation learning, using large demonstration datasets, has become a prominent area of interest in robot learning. The efficacy of imitation learning is heavily reliant on the quantity and quality of the demonstration datasets. In this study, we aim to scale up demonstrations in a data-efficient way to facilitate the learning of generalist robotic agents. We introduce AdaDemo (Adaptive Online Demonstration Expansion), a general framework designed to improve multi-task policy learning by actively and continually expanding the demonstration dataset. AdaDemo strategically collects new demonstrations to address the identified weakness in the existing policy, ensuring data efficiency is maximized. Through a comprehensive evaluation on a total of 22 tasks across two robotic manipulation benchmarks (RLBench and Adroit), we demonstrate AdaDemo's capability to progressively improve policy performance by guiding the generation of high-quality demonstration datasets in a data-efficient manner.