Transferring Hierarchical Structures with Dual Meta Imitation Learning

Chongkai Gao, Yizhou Jiang, Feng Chen
Proceedings of The 6th Conference on Robot Learning, PMLR 205:762-773, 2023.

Abstract

Hierarchical Imitation Learning (HIL) is an effective way for robots to learn sub-skills from long-horizon unsegmented demonstrations. However, the learned hierarchical structure lacks the mechanism to transfer across multi-tasks or to new tasks, which makes them have to learn from scratch when facing a new situation. Transferring and reorganizing modular sub-skills require fast adaptation ability of the whole hierarchical structure. In this work, we propose Dual Meta Imitation Learning (DMIL), a hierarchical meta imitation learning method where the high-level network and sub-skills are iteratively meta-learned with model-agnostic meta-learning. DMIL uses the likelihood of state-action pairs from each sub-skill as the supervision for the high-level network adaptation and uses the adapted high-level network to determine different data set for each sub-skill adaptation. We theoretically prove the convergence of the iterative training process of DMIL and establish the connection between DMIL and Expectation-Maximization algorithm. Empirically, we achieve state-of-the-art few-shot imitation learning performance on the Meta-world benchmark and competitive results on long-horizon tasks in Kitchen environments.

Cite this Paper

BibTeX
@InProceedings{pmlr-v205-gao23b, title = {Transferring Hierarchical Structures with Dual Meta Imitation Learning}, author = {Gao, Chongkai and Jiang, Yizhou and Chen, Feng}, booktitle = {Proceedings of The 6th Conference on Robot Learning}, pages = {762--773}, year = {2023}, editor = {Liu, Karen and Kulic, Dana and Ichnowski, Jeff}, volume = {205}, series = {Proceedings of Machine Learning Research}, month = {14--18 Dec}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v205/gao23b/gao23b.pdf}, url = {https://proceedings.mlr.press/v205/gao23b.html}, abstract = {Hierarchical Imitation Learning (HIL) is an effective way for robots to learn sub-skills from long-horizon unsegmented demonstrations. However, the learned hierarchical structure lacks the mechanism to transfer across multi-tasks or to new tasks, which makes them have to learn from scratch when facing a new situation. Transferring and reorganizing modular sub-skills require fast adaptation ability of the whole hierarchical structure. In this work, we propose Dual Meta Imitation Learning (DMIL), a hierarchical meta imitation learning method where the high-level network and sub-skills are iteratively meta-learned with model-agnostic meta-learning. DMIL uses the likelihood of state-action pairs from each sub-skill as the supervision for the high-level network adaptation and uses the adapted high-level network to determine different data set for each sub-skill adaptation. We theoretically prove the convergence of the iterative training process of DMIL and establish the connection between DMIL and Expectation-Maximization algorithm. Empirically, we achieve state-of-the-art few-shot imitation learning performance on the Meta-world benchmark and competitive results on long-horizon tasks in Kitchen environments.} }
Endnote
%0 Conference Paper %T Transferring Hierarchical Structures with Dual Meta Imitation Learning %A Chongkai Gao %A Yizhou Jiang %A Feng Chen %B Proceedings of The 6th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2023 %E Karen Liu %E Dana Kulic %E Jeff Ichnowski %F pmlr-v205-gao23b %I PMLR %P 762--773 %U https://proceedings.mlr.press/v205/gao23b.html %V 205 %X Hierarchical Imitation Learning (HIL) is an effective way for robots to learn sub-skills from long-horizon unsegmented demonstrations. However, the learned hierarchical structure lacks the mechanism to transfer across multi-tasks or to new tasks, which makes them have to learn from scratch when facing a new situation. Transferring and reorganizing modular sub-skills require fast adaptation ability of the whole hierarchical structure. In this work, we propose Dual Meta Imitation Learning (DMIL), a hierarchical meta imitation learning method where the high-level network and sub-skills are iteratively meta-learned with model-agnostic meta-learning. DMIL uses the likelihood of state-action pairs from each sub-skill as the supervision for the high-level network adaptation and uses the adapted high-level network to determine different data set for each sub-skill adaptation. We theoretically prove the convergence of the iterative training process of DMIL and establish the connection between DMIL and Expectation-Maximization algorithm. Empirically, we achieve state-of-the-art few-shot imitation learning performance on the Meta-world benchmark and competitive results on long-horizon tasks in Kitchen environments.
APA
Gao, C., Jiang, Y. & Chen, F.. (2023). Transferring Hierarchical Structures with Dual Meta Imitation Learning. Proceedings of The 6th Conference on Robot Learning, in Proceedings of Machine Learning Research 205:762-773 Available from https://proceedings.mlr.press/v205/gao23b.html.

Related Material