Conference paper Open Access

Cluster Variatonal Approximations for Structure Learning of Continuous-Time Bayesian Networks from Incomplete Data

LInzner, Dominik; Koeppl, Heinz

Continuous-time Bayesian networks (CTBNs) constitute a general and powerful framework for modeling continuous-time stochastic processes on networks. This makes them particularly attractive for learning the directed structures among interacting entities. However, if the available data is incomplete, one needs to simulate the prohibitively complex CTBN dynamics. Existing approximation techniques, such as sampling and low-order variational methods, either scale unfavorably in system size, or are unsatisfactory in terms of accuracy. Inspired by recent advances in statistical physics, we present a new approximation scheme based on cluster-variational methods significantly improving upon existing variational approximations.  We can analytically marginalize the parameters of the approximate CTBN, as these are of secondary importance for structure learning. This recovers a
scalable scheme for direct structure learning from incomplete and noisy time-series data. Our approach outperforms existing methods in terms of scalability.

Files (2.3 MB)
Name Size
cvm_ctbn_nips.pdf
md5:f96b2076ad034b312f962c667187ca9c
2.3 MB Download
32
34
views
downloads
All versions This version
Views 3232
Downloads 3434
Data volume 77.0 MB77.0 MB
Unique views 3131
Unique downloads 3333

Share

Cite as