Learning Causal Dynamics Models in Object-Oriented Environments

Zhongwei Yu, Jingqing Ruan, Dengpeng Xing

Primary Area: reinforcement learning

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: reinforcement learning, causality, dynamics model

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

TL;DR: Improve the scalability of causal dynamics models in object-oriented environments.

Abstract: Causal Dynamics Models (CDMs) have demonstrated significant potential in addressing various challenges in reinforcement learning. Recent studies have incorporated causal discovery to capture the causal dependencies among environmental variables in the learning of CDMs. 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. Experimental results from large-scale tasks indicate that OOCDM outperforms existing CDMs in terms of causal discovery, prediction accuracy, generalization, and computational efficiency.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 4429