Bayesian Nonparametric Methods for Learning Markov Switching Processes
In this article, we explored a Bayesian nonparametric approach to learning Markov switching processes. This framework requires one to make fewer assumptions about the underlying dynamics, and thereby allows the data to drive the complexity of the inferred model. We began by examining a Bayesian nonp...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2012
|
| Online Access: | http://hdl.handle.net/1721.1/69030 https://orcid.org/0000-0003-0149-5888 |
| _version_ | 1826189709146062848 |
|---|---|
| author | Fox, Emily Beth Willsky, Alan S. Sudderth, Erik B. Jordan, Michael I. |
| 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 Willsky, Alan S. Sudderth, Erik B. Jordan, Michael I. |
| author_sort | Fox, Emily Beth |
| collection | MIT |
| description | In this article, we explored a Bayesian nonparametric approach to learning Markov switching processes. This framework requires one to make fewer assumptions about the underlying dynamics, and thereby allows the data to drive the complexity of the inferred model. We began by examining a Bayesian nonparametric HMM, the sticky HDPHMM, that uses a hierarchical DP prior to regularize an unbounded mode space. We then considered extensions to Markov switching processes with richer, conditionally linear dynamics, including the HDP-AR-HMM and HDP-SLDS. We concluded by considering methods for transferring knowledge among multiple related time series. We argued that a featural representation is more appropriate than a rigid global clustering, as it encourages sharing of behaviors among objects while still allowing sequence-specific variability. In this context, the beta process provides an appealing alternative to the DP. |
| first_indexed | 2024-09-23T08:19:48Z |
| format | Article |
| id | mit-1721.1/69030 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T08:19:48Z |
| publishDate | 2012 |
| publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| record_format | dspace |
| spelling | mit-1721.1/690302022-09-23T12:20:48Z Bayesian Nonparametric Methods for Learning Markov Switching Processes Fox, Emily Beth Willsky, Alan S. Sudderth, Erik B. Jordan, Michael I. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Willskey, Alan S. Fox, Emily Beth Willsky, Alan S. In this article, we explored a Bayesian nonparametric approach to learning Markov switching processes. This framework requires one to make fewer assumptions about the underlying dynamics, and thereby allows the data to drive the complexity of the inferred model. We began by examining a Bayesian nonparametric HMM, the sticky HDPHMM, that uses a hierarchical DP prior to regularize an unbounded mode space. We then considered extensions to Markov switching processes with richer, conditionally linear dynamics, including the HDP-AR-HMM and HDP-SLDS. We concluded by considering methods for transferring knowledge among multiple related time series. We argued that a featural representation is more appropriate than a rigid global clustering, as it encourages sharing of behaviors among objects while still allowing sequence-specific variability. In this context, the beta process provides an appealing alternative to the DP. 2012-02-03T19:29:20Z 2012-02-03T19:29:20Z 2010-11 Article http://purl.org/eprint/type/JournalArticle 1053-5888 INSPEC Accession Number: 11588731 http://hdl.handle.net/1721.1/69030 Fox, Emily et al. “Bayesian Nonparametric Methods for Learning Markov Switching Processes.” IEEE Signal Processing Magazine (2010): n. pag. Web. 3 Feb. 2012. © 2011 Institute of Electrical and Electronics Engineers https://orcid.org/0000-0003-0149-5888 en_US http://dx.doi.org/10.1109/msp.2010.937999 IEEE Signal Processing Magazine Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/pdf Institute of Electrical and Electronics Engineers (IEEE) IEEE |
| spellingShingle | Fox, Emily Beth Willsky, Alan S. Sudderth, Erik B. Jordan, Michael I. Bayesian Nonparametric Methods for Learning Markov Switching Processes |
| title | Bayesian Nonparametric Methods for Learning Markov Switching Processes |
| title_full | Bayesian Nonparametric Methods for Learning Markov Switching Processes |
| title_fullStr | Bayesian Nonparametric Methods for Learning Markov Switching Processes |
| title_full_unstemmed | Bayesian Nonparametric Methods for Learning Markov Switching Processes |
| title_short | Bayesian Nonparametric Methods for Learning Markov Switching Processes |
| title_sort | bayesian nonparametric methods for learning markov switching processes |
| url | http://hdl.handle.net/1721.1/69030 https://orcid.org/0000-0003-0149-5888 |
| work_keys_str_mv | AT foxemilybeth bayesiannonparametricmethodsforlearningmarkovswitchingprocesses AT willskyalans bayesiannonparametricmethodsforlearningmarkovswitchingprocesses AT suddertherikb bayesiannonparametricmethodsforlearningmarkovswitchingprocesses AT jordanmichaeli bayesiannonparametricmethodsforlearningmarkovswitchingprocesses |