Exchangeable Neural ODE for Set Modeling
This work addresses a fundamental challenge in set modeling for applications like point cloud processing, offering a novel approach to improve permutation equivariance.
The authors tackled the problem of modeling unordered sets with permutation equivariance by proposing Exchangeable Neural ODE (ExNODE), a method based on ordinary differential equations that achieves strong performance in discriminative and generative tasks, as shown in extensive experiments.
Reasoning over an instance composed of a set of vectors, like a point cloud, requires that one accounts for intra-set dependent features among elements. However, since such instances are unordered, the elements' features should remain unchanged when the input's order is permuted. This property, permutation equivariance, is a challenging constraint for most neural architectures. While recent work has proposed global pooling and attention-based solutions, these may be limited in the way that intradependencies are captured in practice. In this work we propose a more general formulation to achieve permutation equivariance through ordinary differential equations (ODE). Our proposed module, Exchangeable Neural ODE (ExNODE), can be seamlessly applied for both discriminative and generative tasks. We also extend set modeling in the temporal dimension and propose a VAE based model for temporal set modeling. Extensive experiments demonstrate the efficacy of our method over strong baselines.