Search papers, labs, and topics across Lattice.
This paper introduces a hybrid neural-physics framework that integrates known components of ordinary differential equations (ODEs) with neural networks to model dynamical systems when some equations are unknown and only partial state measurements are available. The method alternates between state estimation using a Rauch-Tung-Striebel (RTS) smoother and parameter estimation for the neural network, iterating until convergence. The results demonstrate that this approach effectively learns missing ODE components while enhancing latent-state reconstruction and long-horizon predictions across various dynamical systems.
Missing components of dynamical systems can be accurately learned from partial observations, enhancing predictions without sacrificing interpretability.
Ordinary differential equations (ODEs) are widely used to model dynamical systems in physics, biology, neuroscience, and physiology, but in many applications some equations of the dynamics are unknown and only a subset of the state variables are measured. We propose a hybrid neural--physics framework in which the known components of the ODE are kept explicit and the missing components are represented by a neural network. The proposed method consists of two stages where we alternate between state and parameter estimation and iterate until a predetermined criterion is met. Specifically, in the first step, we treat the model parameters as being known and we infer the latent states from the available measurements using a Rauch--Tung--Striebel (RTS) smoother. In the second stage, we treat the smoothed trajectories as being known and use them to estimate the neural networks'parameters through backpropagation. We evaluate the method on benchmark systems spanning linear, nonlinear, and stiff dynamics under partial state observation. Across these settings, the proposed method learns missing ODE components from incomplete measurements while exploiting and retaining interpretable mechanistic structure and improving latent-state reconstruction and long-horizon prediction.
