Integrating Symmetry into Differentiable Planning with Steerable ConvolutionsDownload PDF


22 Sept 2022, 12:39 (modified: 16 Nov 2022, 19:30)ICLR 2023 Conference Blind SubmissionReaders: Everyone
Abstract: We study how group symmetry helps improve data efficiency and generalization for end-to-end differentiable planning algorithms, when symmetry appears in decision-making tasks. Motivated by equivariant convolution networks, we treat the path planning problem as \textit{signals} over grids. We show that value iteration in this case is a \textit{linear equivariant operator}, which is a (steerable) \textit{convolution}. This extends Value Iteration Networks (VINs) on using convolutional networks for path planning with additional \textit{rotation} and \textit{reflection} symmetry. Our implementation is based on VINs and uses steerable convolution networks to incorporate symmetry. The experiments are performed on four tasks: 2D navigation, visual navigation, 2 degrees of freedom (2DOFs) configuration space and workspace manipulation. % in configuration space or workspace. Our symmetric planning algorithms improve training efficiency and generalization by large margins compared to non-equivariant counterparts, VIN and GPPN.
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.
Supplementary Material: zip
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
Submission Guidelines: Yes
Please Choose The Closest Area That Your Submission Falls Into: Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)
8 Replies