Name
Abstract | ||
|---|---|---|
We consider the problem of Bayesian inference of graphical structure describing the interactions among multiple vector time-series. A directed temporal interaction model is presented which assumes a fixed dependence structure among time-series. Using a conjugate prior over this model's structure and parameters, we focus our attention on characterizing the exact posterior uncertainty in the structure given data. The model is extended via the introduction of a dy- namically evolving latent variable which indexes dependence structures over time. Performing in- ference using this model yields promising re- sults when analyzing the interaction of multiple tracked moving objects. |
Year | Venue | Keywords |
|---|---|---|
2009 | AISTATS | indexation,latent variable,time series,conjugate prior,bayesian inference |
Field | DocType | Volume |
Frequentist inference,Bayesian inference,Fiducial inference,Inference,Computer science,Bayesian linear regression,Latent variable,Artificial intelligence,Bayesian statistics,Conjugate prior,Machine learning | Journal | 5 |
Citations | PageRank | References |
6 | 0.63 | 11 |
Authors | ||
2 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Michael R. Siracusa | 1 | 13 | 1.62 |
| John W. Fisher III | 2 | 878 | 74.44 |