Topology Learning of Linear Dynamical Systems with Latent Nodes using Matrix Decomposition

16 Dec 2019  ·  Mishfad S. V., Harish Doddi, Murti V. Salapaka ·

In this article, we present a novel approach to reconstruct the topology of networked linear dynamical systems with latent nodes. The network is allowed to have directed loops and bi-directed edges. The main approach relies on the unique decomposition of the inverse of power spectral density matrix (IPSDM) obtained from observed nodes as a sum of sparse and low-rank matrices. We provide conditions and methods for decomposing the IPSDM of the observed nodes into sparse and low-rank components. The sparse component yields the moral graph associated with the observed nodes, and the low-rank component retrieves parents, children and spouses (the Markov Blanket) of the hidden nodes. The article provides necessary and sufficient conditions for the unique decomposition of a given skew symmetric matrix into sum of a sparse skew symmetric and a low-rank skew symmetric matrices. It is shown that for a large class of systems, the unique decomposition of imaginary part of the IPSDM of observed nodes, a skew symmetric matrix, into the sparse and the low-rank components is sufficient to identify the moral graph of the observed nodes as well as the Markov Blanket of latent nodes. For a large class of systems, all spurious links in the moral graph formed by the observed nodes can be identified. Assuming conditions on hidden nodes required for identifiability, links between the hidden and observed nodes can be reconstructed, resulting in the retrieval of the exact topology of the network from the availability of IPSDM. Moreover, for finite number of data samples, we provide concentration bounds on the entry-wise distance between the true IPSDM and the estimated IPSDM.

PDF Abstract
No code implementations yet. Submit your code now

Datasets


  Add Datasets introduced or used in this paper

Results from the Paper


  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

Methods


No methods listed for this paper. Add relevant methods here