Towards Geometric Normalization Techniques in SE(3) Equivariant Graph Neural Networks for Physical Dynamics Simulations
Abstract
The abstract visual reasoning (AVR) domain presents a diverse suite of analogy-based tasks devoted to studying model generalization. Recent years have brought dynamic progress in the field, particularly in i.i.d. scenarios, in which models are trained and evaluated on the same data distributions. Nevertheless, o.o.d. setups that assess model generalization to new test distributions remain challenging even for the most recent models. To advance generalization in AVR tasks, we present the Pathways of Normalized Group Convolution model (PoNG), a novel neural architecture that features group convolution, normalization, and a parallel design. We consider a wide set of AVR benchmarks, including Raven's Progressive Matrices and visual analogy problems with both synthetic and real-world images. The experiments demonstrate strong generalization capabilities of the proposed model, which in several settings outperforms the existing literature methods.
Cite
Text
Meng et al. "Towards Geometric Normalization Techniques in SE(3) Equivariant Graph Neural Networks for Physical Dynamics Simulations." International Joint Conference on Artificial Intelligence, 2024. doi:10.24963/ijcai.2024/661Markdown
[Meng et al. "Towards Geometric Normalization Techniques in SE(3) Equivariant Graph Neural Networks for Physical Dynamics Simulations." International Joint Conference on Artificial Intelligence, 2024.](https://mlanthology.org/ijcai/2024/meng2024ijcai-geometric/) doi:10.24963/ijcai.2024/661BibTeX
@inproceedings{meng2024ijcai-geometric,
title = {{Towards Geometric Normalization Techniques in SE(3) Equivariant Graph Neural Networks for Physical Dynamics Simulations}},
author = {Meng, Ziqiao and Zeng, Liang and Song, Zixing and Xu, Tingyang and Zhao, Peilin and King, Irwin},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2024},
pages = {5981-5989},
doi = {10.24963/ijcai.2024/661},
url = {https://mlanthology.org/ijcai/2024/meng2024ijcai-geometric/}
}