Visual Goal-Directed Meta-Imitation Learning

Corban G. Rivera, David A. Handelman, Christopher R. Ratto, David Patrone, Bart L. Paulhamus; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2022, pp. 3767-3773


The goal of meta-learning is to generalize to new tasks and goals as quickly as possible. Ideally, we would like approaches that generalize to new goals and tasks on the first attempt. Requiring a policy to perform on a new task on the first attempt without even a single example trajectory is a zero-shot problem formulation. When tasks are identified by goal images, the tasks can be considered visually goal-directed. In this work, we explore the problem of visual goal-directed zero-shot meta-imitation learning. Inspired by several popular approaches to Meta-RL, we composed several core ideas related to task-embedding and planning by gradient descent to attempt to explore this problem. To evaluate these approaches, we adapted the Metaworld benchmark tasks to create 24 distinct visual goal-directed manipulation tasks. We found that 7 out of 24 tasks could be successfully completed on the first attempt by at least one of the approaches we tested. We demonstrated that goal-directed zero-shot approaches can translate to a physical robot with a demonstration based on Jenga block manipulation tasks using a Kinova Jaco robotic arm.

Related Material

[pdf] [supp]
@InProceedings{Rivera_2022_CVPR, author = {Rivera, Corban G. and Handelman, David A. and Ratto, Christopher R. and Patrone, David and Paulhamus, Bart L.}, title = {Visual Goal-Directed Meta-Imitation Learning}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2022}, pages = {3767-3773} }