[1B4-OS-41b] OS-41

[1B4-OS-41b-05] On the Robustness of Object-Centric Representations for Model-Based Reinforcement Learning

〇Akihiro Nakano¹, Masahiro Suzuki¹, Yutaka Matsuo¹ (1. The University of Tokyo)

Keywords:World models, Object-centric learning, Robustness

Model-based reinforcement learning (RL) is a promising approach to learning to control agents in a sample-efficient manner, but often struggles with generalization beyond tasks it was trained on. While previous work have explored using pretrained visual representations (PVR) to improve generalization, these approaches have not outperformed representations learned from scratch in out-of-distribution (OOD) settings. In this work, we propose to incorporate object-centric representations, which have demonstrated strong OOD generalization capabilities by learning compositional representations, into model-based RL with PVR. We investigate whether this object-centric inductive bias improves both sample efficiency and task performance across in-distribution and OOD environments.