Learning Causal Dynamics Models in Object-Oriented Environments

Zhongwei Yu, Jingqing Ruan, Dengpeng Xing
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:57597-57638, 2024.

Abstract

Causal dynamics models (CDMs) have demonstrated significant potential in addressing various challenges in reinforcement learning. To learn CDMs, recent studies have performed causal discovery to capture the causal dependencies among environmental variables. However, the learning of CDMs is still confined to small-scale environments due to computational complexity and sample efficiency constraints. This paper aims to extend CDMs to large-scale object-oriented environments, which consist of a multitude of objects classified into different categories. We introduce the Object-Oriented CDM (OOCDM) that shares causalities and parameters among objects belonging to the same class. Furthermore, we propose a learning method for OOCDM that enables it to adapt to a varying number of objects. Experiments on large-scale tasks indicate that OOCDM outperforms existing CDMs in terms of causal discovery, prediction accuracy, generalization, and computational efficiency.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-yu24j, title = {Learning Causal Dynamics Models in Object-Oriented Environments}, author = {Yu, Zhongwei and Ruan, Jingqing and Xing, Dengpeng}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {57597--57638}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24j/yu24j.pdf}, url = {https://proceedings.mlr.press/v235/yu24j.html}, abstract = {Causal dynamics models (CDMs) have demonstrated significant potential in addressing various challenges in reinforcement learning. To learn CDMs, recent studies have performed causal discovery to capture the causal dependencies among environmental variables. However, the learning of CDMs is still confined to small-scale environments due to computational complexity and sample efficiency constraints. This paper aims to extend CDMs to large-scale object-oriented environments, which consist of a multitude of objects classified into different categories. We introduce the Object-Oriented CDM (OOCDM) that shares causalities and parameters among objects belonging to the same class. Furthermore, we propose a learning method for OOCDM that enables it to adapt to a varying number of objects. Experiments on large-scale tasks indicate that OOCDM outperforms existing CDMs in terms of causal discovery, prediction accuracy, generalization, and computational efficiency.} }
Endnote
%0 Conference Paper %T Learning Causal Dynamics Models in Object-Oriented Environments %A Zhongwei Yu %A Jingqing Ruan %A Dengpeng Xing %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-yu24j %I PMLR %P 57597--57638 %U https://proceedings.mlr.press/v235/yu24j.html %V 235 %X Causal dynamics models (CDMs) have demonstrated significant potential in addressing various challenges in reinforcement learning. To learn CDMs, recent studies have performed causal discovery to capture the causal dependencies among environmental variables. However, the learning of CDMs is still confined to small-scale environments due to computational complexity and sample efficiency constraints. This paper aims to extend CDMs to large-scale object-oriented environments, which consist of a multitude of objects classified into different categories. We introduce the Object-Oriented CDM (OOCDM) that shares causalities and parameters among objects belonging to the same class. Furthermore, we propose a learning method for OOCDM that enables it to adapt to a varying number of objects. Experiments on large-scale tasks indicate that OOCDM outperforms existing CDMs in terms of causal discovery, prediction accuracy, generalization, and computational efficiency.
APA
Yu, Z., Ruan, J. & Xing, D.. (2024). Learning Causal Dynamics Models in Object-Oriented Environments. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:57597-57638 Available from https://proceedings.mlr.press/v235/yu24j.html.

Related Material