SCALE: Causal Learning and Discovery of Robot Manipulation Skills using Simulation

Tabitha Edith Lee, Shivam Vats, Siddharth Girdhar, Oliver Kroemer
Proceedings of The 7th Conference on Robot Learning, PMLR 229:2229-2256, 2023.

Abstract

We propose SCALE, an approach for discovering and learning a diverse set of interpretable robot skills from a limited dataset. Rather than learning a single skill which may fail to capture all the modes in the data, we first identify the different modes via causal reasoning and learn a separate skill for each of them. Our main insight is to associate each mode with a unique set of causally relevant context variables that are discovered by performing causal interventions in simulation. This enables data partitioning based on the causal processes that generated the data, and then compressed skills that ignore the irrelevant variables can be trained. We model each robot skill as a Regional Compressed Option, which extends the options framework by associating a causal process and its relevant variables with the option. Modeled as the skill Data Generating Region, each causal process is local in nature and hence valid over only a subset of the context space. We demonstrate our approach for two representative manipulation tasks: block stacking and peg-in-hole insertion under uncertainty. Our experiments show that our approach yields diverse skills that are compact, robust to domain shifts, and suitable for sim-to-real transfer.

Cite this Paper

BibTeX
@InProceedings{pmlr-v229-lee23b, title = {SCALE: Causal Learning and Discovery of Robot Manipulation Skills using Simulation}, author = {Lee, Tabitha Edith and Vats, Shivam and Girdhar, Siddharth and Kroemer, Oliver}, booktitle = {Proceedings of The 7th Conference on Robot Learning}, pages = {2229--2256}, year = {2023}, editor = {Tan, Jie and Toussaint, Marc and Darvish, Kourosh}, volume = {229}, series = {Proceedings of Machine Learning Research}, month = {06--09 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v229/lee23b/lee23b.pdf}, url = {https://proceedings.mlr.press/v229/lee23b.html}, abstract = {We propose SCALE, an approach for discovering and learning a diverse set of interpretable robot skills from a limited dataset. Rather than learning a single skill which may fail to capture all the modes in the data, we first identify the different modes via causal reasoning and learn a separate skill for each of them. Our main insight is to associate each mode with a unique set of causally relevant context variables that are discovered by performing causal interventions in simulation. This enables data partitioning based on the causal processes that generated the data, and then compressed skills that ignore the irrelevant variables can be trained. We model each robot skill as a Regional Compressed Option, which extends the options framework by associating a causal process and its relevant variables with the option. Modeled as the skill Data Generating Region, each causal process is local in nature and hence valid over only a subset of the context space. We demonstrate our approach for two representative manipulation tasks: block stacking and peg-in-hole insertion under uncertainty. Our experiments show that our approach yields diverse skills that are compact, robust to domain shifts, and suitable for sim-to-real transfer.} }
Endnote
%0 Conference Paper %T SCALE: Causal Learning and Discovery of Robot Manipulation Skills using Simulation %A Tabitha Edith Lee %A Shivam Vats %A Siddharth Girdhar %A Oliver Kroemer %B Proceedings of The 7th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2023 %E Jie Tan %E Marc Toussaint %E Kourosh Darvish %F pmlr-v229-lee23b %I PMLR %P 2229--2256 %U https://proceedings.mlr.press/v229/lee23b.html %V 229 %X We propose SCALE, an approach for discovering and learning a diverse set of interpretable robot skills from a limited dataset. Rather than learning a single skill which may fail to capture all the modes in the data, we first identify the different modes via causal reasoning and learn a separate skill for each of them. Our main insight is to associate each mode with a unique set of causally relevant context variables that are discovered by performing causal interventions in simulation. This enables data partitioning based on the causal processes that generated the data, and then compressed skills that ignore the irrelevant variables can be trained. We model each robot skill as a Regional Compressed Option, which extends the options framework by associating a causal process and its relevant variables with the option. Modeled as the skill Data Generating Region, each causal process is local in nature and hence valid over only a subset of the context space. We demonstrate our approach for two representative manipulation tasks: block stacking and peg-in-hole insertion under uncertainty. Our experiments show that our approach yields diverse skills that are compact, robust to domain shifts, and suitable for sim-to-real transfer.
APA
Lee, T.E., Vats, S., Girdhar, S. & Kroemer, O.. (2023). SCALE: Causal Learning and Discovery of Robot Manipulation Skills using Simulation. Proceedings of The 7th Conference on Robot Learning, in Proceedings of Machine Learning Research 229:2229-2256 Available from https://proceedings.mlr.press/v229/lee23b.html.

Related Material