Using Latent Binary Variables for Online Reconstruction of Large Scale Systems
arXiv:1312.6607 · doi:10.1007/s10472-015-9470-x
Abstract
We propose a probabilistic graphical model realizing a minimal encoding of real variables dependencies based on possibly incomplete observation and an empirical cumulative distribution function per variable. The target application is a large scale partially observed system, like e.g. a traffic network, where a small proportion of real valued variables are observed, and the other variables have to be predicted. Our design objective is therefore to have good scalability in a real-time setting. Instead of attempting to encode the dependencies of the system directly in the description space, we propose a way to encode them in a latent space of binary variables, reflecting a rough perception of the observable (congested/non-congested for a traffic road). The method relies in part on message passing algorithms, i.e. belief propagation, but the core of the work concerns the definition of meaningful latent variables associated to the variables of interest and their pairwise dependencies. Numerical experiments demonstrate the applicability of the method in practice.
References in corpus (5)
- High-dimensional Ising model selection using -regularized logistic regression
- Sufficient conditions for convergence of the Sum-Product Algorithm
- Adaptive cluster expansion for the inverse Ising problem: convergence, algorithm and tests
- Gaussian Belief Propagation Based Multiuser Detection
- Learning Multiple Belief Propagation Fixed Points for Real Time Inference