EDGI: Equivariant Diffusion for Planning with Embodied AgentsDownload PDF

Published: 03 Mar 2023, Last Modified: 12 Apr 2023RRL 2023 PosterReaders: Everyone
Keywords: Diffuser, Diffusion Models, Offline RL, Planning, Equivariance
TL;DR: We introduce a planning algorithm for embodied agents, based on a SE(3) × ℤ × Sₙ-equivariant diffusion model.
Abstract: Embodied agents operate in a structured world, often solving tasks with spatial, temporal, and permutation symmetries. Most algorithms for planning and model-based reinforcement learning (MBRL) do not take this rich geometric structure into account, leading to sample inefficiency and poor generalization. We introduce the Equivariant Diffuser for Generating Interactions (EDGI), an algorithm for MBRL and planning that is equivariant with respect to the product of the spatial symmetry group SE(3), the discrete-time translation group ℤ, and the object permutation group Sₙ. EDGI follows the Diffuser framework (Janner et al. 2022) in treating both learning a world model and planning in it as a conditional generative modeling problem, training a diffusion model on an offline trajectory dataset. We introduce a new SE(3) × ℤ × Sₙ-equivariant diffusion model that supports multiple representations. We integrate this model in a planning loop, where conditioning and classifier-based guidance allow us to softly break the symmetry for specific tasks as needed. On navigation and object manipulation tasks, EDGI improves sample efficiency and generalization.
Track: Technical Paper
Confirmation: I have read and agree with the workshop's policy on behalf of myself and my co-authors.
2 Replies