Neural Modes: Self-supervised Learning of Nonlinear Modal Subspaces

arXiv:2404.17620v1 Announce Type: new
Abstract: We propose a self-supervised approach for learning physics-based subspaces for real-time simulation. Existing learning-based methods construct subspaces by approximating pre-defined simulation data in a purely geometric way. However, this approach tends to produce high-energy configurations, leads to entangled latent space dimensions, and generalizes poorly beyond the training set. To overcome these limitations, we propose a self-supervised approach that directly minimizes the system's mechanical energy during training. We show that our method leads to learned subspaces …

abstract arxiv beyond construct cs.cv cs.gr cs.lg data dimensions energy however leads modal physics real-time self-supervised learning simulation space supervised learning training type