Abstract | ||
---|---|---|
Many nonlinear dynamical phenomena can be effectively modeled by a system that switches among a set of conditionally linear dynamical modes. We con- sider two such models: the switching linear dynamical system (SLDS) and the switching vector autoregressive (VAR) process. Our nonparametric Bayesian ap- proach utilizes a hierarchical Dirichlet process prior to l earn an unknown number of persistent, smooth dynamical modes. We develop a sampling algorithm that combines a truncated approximation to the Dirichlet process with efficient joint sampling of the mode and state sequences. The utility and flex ibility of our model are demonstrated on synthetic data, sequences of dancing honey bees, and the IBOVESPA stock index. |
Year | Venue | Keywords |
---|---|---|
2008 | NIPS | indexation,synthetic data,vector autoregression,nonlinear dynamics,hierarchical dirichlet process,linear dynamical system,bayesian learning |
Field | DocType | Citations |
Autoregressive model,Hierarchical Dirichlet process,Linear dynamical system,Mathematical optimization,Dirichlet process,Nonlinear system,Computer science,Synthetic data,Sampling (statistics),Artificial intelligence,Random dynamical system,Machine learning | Conference | 53 |
PageRank | References | Authors |
2.02 | 10 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Emily B. Fox | 1 | 542 | 39.91 |
Erik B. Sudderth | 2 | 1420 | 119.04 |
Michael I. Jordan | 3 | 31220 | 3640.80 |
Alan S. Willsky | 4 | 7466 | 847.01 |