Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Embedding Hybrid Systems into Continuous Latent Vector Fields

About

This work proves that an $n$-dimensional hybrid system can be embedded into an $m$-dimensional Euclidean space equipped with a continuous vector field on its embedded image whenever $m>2n$. This result suggests that an intrinsically discontinuous hybrid system generically admits a continuous extrinsic representation that is well-posed for differentiable optimization. Building on this existence theorem, we show that a latent Neural ODE with consistency loss in both the latent and state space can accurately recover the flow of hybrid systems. Extensive experiments suggest the proposed method outperforms the existing method in learning hybrid systems with varying geometries from only time series data.

Sangli Teng, Hang Liu, Koushil Sreenath• 2026

Related benchmarks

TaskDatasetResultRank
Learning Dynamical SystemsThree-Link Walker time-series
MSE0.0952
4
Learning Dynamical SystemsBouncing Ball time-series
MSE0.158
2
Learning Dynamical SystemsTorus time-series
MSE0.0037
2
Learning Dynamical SystemsKlein Bottle time-series
MSE0.0059
2
Learning Dynamical Systems3D Bouncing Ball
MSE0.162
2
Showing 5 of 5 rows

Other info

Follow for update