Bayesian Nonparametric Inference of Switching Dynamic Linear Models
Many complex dynamical phenomena can be effectively modeled by a system that switches among a set of conditionally linear dynamical modes. We consider two such models: the switching linear dynamical system (SLDS) and the switching vector autoregressive (VAR) process. Our Bayesian nonparametric appro...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2013
|
| Online Access: | http://hdl.handle.net/1721.1/80811 https://orcid.org/0000-0003-0149-5888 |
| _version_ | 1826196404042727424 |
|---|---|
| author | Fox, Emily Beth Sudderth, Erik B. Jordan, Michael I. Willsky, Alan S. |
| author2 | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science |
| author_facet | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Fox, Emily Beth Sudderth, Erik B. Jordan, Michael I. Willsky, Alan S. |
| author_sort | Fox, Emily Beth |
| collection | MIT |
| description | Many complex dynamical phenomena can be effectively modeled by a system that switches among a set of conditionally linear dynamical modes. We consider two such models: the switching linear dynamical system (SLDS) and the switching vector autoregressive (VAR) process. Our Bayesian nonparametric approach utilizes a hierarchical Dirichlet process prior to learn an unknown number of persistent, smooth dynamical modes. We additionally employ automatic relevance determination to infer a sparse set of dynamic dependencies allowing us to learn SLDS with varying state dimension or switching VAR processes with varying autoregressive order. 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 flexibility of our model are demonstrated on synthetic data, sequences of dancing honey bees, the IBOVESPA stock index and a maneuvering target tracking application. |
| first_indexed | 2024-09-23T10:26:18Z |
| format | Article |
| id | mit-1721.1/80811 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T10:26:18Z |
| publishDate | 2013 |
| publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| record_format | dspace |
| spelling | mit-1721.1/808112022-09-26T17:53:10Z Bayesian Nonparametric Inference of Switching Dynamic Linear Models Fox, Emily Beth Sudderth, Erik B. Jordan, Michael I. Willsky, Alan S. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Willsky, Alan S. Many complex dynamical phenomena can be effectively modeled by a system that switches among a set of conditionally linear dynamical modes. We consider two such models: the switching linear dynamical system (SLDS) and the switching vector autoregressive (VAR) process. Our Bayesian nonparametric approach utilizes a hierarchical Dirichlet process prior to learn an unknown number of persistent, smooth dynamical modes. We additionally employ automatic relevance determination to infer a sparse set of dynamic dependencies allowing us to learn SLDS with varying state dimension or switching VAR processes with varying autoregressive order. 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 flexibility of our model are demonstrated on synthetic data, sequences of dancing honey bees, the IBOVESPA stock index and a maneuvering target tracking application. United States. Air Force Office of Scientific Research. Multidisciplinary University Research Initiative (Grant FA9550-06-1-0324) United States. Army Research Office (Grant W911NF-06-1-0076) 2013-09-19T18:54:28Z 2013-09-19T18:54:28Z 2011-01 2010-12 Article http://purl.org/eprint/type/JournalArticle 1053-587X 1941-0476 http://hdl.handle.net/1721.1/80811 Fox, Emily, Erik B. Sudderth, Michael I. Jordan, and Alan S. Willsky. Bayesian Nonparametric Inference of Switching Dynamic Linear Models. IEEE Transactions on Signal Processing 59, no. 4 (April 2011): 1569-1585. https://orcid.org/0000-0003-0149-5888 en_US http://dx.doi.org/10.1109/tsp.2010.2102756 IEEE Transactions on Signal Processing Creative Commons Attribution-Noncommercial-Share Alike 3.0 http://creativecommons.org/licenses/by-nc-sa/3.0/ application/pdf Institute of Electrical and Electronics Engineers (IEEE) Willsky via Amy Stout |
| spellingShingle | Fox, Emily Beth Sudderth, Erik B. Jordan, Michael I. Willsky, Alan S. Bayesian Nonparametric Inference of Switching Dynamic Linear Models |
| title | Bayesian Nonparametric Inference of Switching Dynamic Linear Models |
| title_full | Bayesian Nonparametric Inference of Switching Dynamic Linear Models |
| title_fullStr | Bayesian Nonparametric Inference of Switching Dynamic Linear Models |
| title_full_unstemmed | Bayesian Nonparametric Inference of Switching Dynamic Linear Models |
| title_short | Bayesian Nonparametric Inference of Switching Dynamic Linear Models |
| title_sort | bayesian nonparametric inference of switching dynamic linear models |
| url | http://hdl.handle.net/1721.1/80811 https://orcid.org/0000-0003-0149-5888 |
| work_keys_str_mv | AT foxemilybeth bayesiannonparametricinferenceofswitchingdynamiclinearmodels AT suddertherikb bayesiannonparametricinferenceofswitchingdynamiclinearmodels AT jordanmichaeli bayesiannonparametricinferenceofswitchingdynamiclinearmodels AT willskyalans bayesiannonparametricinferenceofswitchingdynamiclinearmodels |