Transfer Learning in Robotics: An Upcoming Breakthrough? A Review of Promises and Challenges

TL;DR

This review paper discusses the potential and challenges of transfer learning in robotics, emphasizing the need for a taxonomy, understanding transfer gaps, and addressing negative transfer to advance intelligent robotic systems.

Contribution

It provides the first taxonomy of transfer learning in robotics, unifies various perspectives, and highlights key challenges and research directions.

Findings

Identifies the need for transfer at different abstraction levels

Highlights the importance of quantifying transfer gaps and transfer quality

Warns about the risks of negative transfer

Abstract

Transfer learning is a conceptually-enticing paradigm in pursuit of truly intelligent embodied agents. The core concept -- reusing prior knowledge to learn in and from novel situations -- is successfully leveraged by humans to handle novel situations. In recent years, transfer learning has received renewed interest from the community from different perspectives, including imitation learning, domain adaptation, and transfer of experience from simulation to the real world, among others. In this paper, we unify the concept of transfer learning in robotics and provide the first taxonomy of its kind considering the key concepts of robot, task, and environment. Through a review of the promises and challenges in the field, we identify the need of transferring at different abstraction levels, the need of quantifying the transfer gap and the quality of transfer, as well as the dangers of negative transfer. Via this position paper, we hope to channel the effort of the community towards the most significant roadblocks to realize the full potential of transfer learning in robotics.