Symmetric Linear Dynamical Systems are Learnable from Few Observations
By: Minh Vu, Andrey Y. Lokhov, Marc Vuffray
We consider the problem of learning the parameters of a $N$-dimensional stochastic linear dynamics under both full and partial observations from a single trajectory of time $T$. We introduce and analyze a new estimator that achieves a small maximum element-wise error on the recovery of symmetric dynamic matrices using only $T=\mathcal{O}(\log N)$ observations, irrespective of whether the matrix is sparse or dense. This estimator is based on the method of moments and does not rely on problem-specific regularization. This is especially important for applications such as structure discovery.
Similar Papers
Universal Learning of Nonlinear Dynamics
Machine Learning (CS)
Learns how things change, even when wobbly.
Latent Mixture of Symmetries for Sample-Efficient Dynamic Learning
Machine Learning (CS)
Teaches robots to learn faster with less data.
Learning-Enhanced Observer for Linear Time-Invariant Systems with Parametric Uncertainty
Machine Learning (CS)
Improves guessing of hidden system parts.