Cutting plane algorithms for variational inference in graphical models
Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2007.
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | eng |
| Published: |
Massachusetts Institute of Technology
2008
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/40327 |
| _version_ | 1826201169992613888 |
|---|---|
| author | Sontag, David Alexander |
| author2 | Tommi S. Jaakkola. |
| author_facet | Tommi S. Jaakkola. Sontag, David Alexander |
| author_sort | Sontag, David Alexander |
| collection | MIT |
| description | Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2007. |
| first_indexed | 2024-09-23T11:47:35Z |
| format | Thesis |
| id | mit-1721.1/40327 |
| institution | Massachusetts Institute of Technology |
| language | eng |
| last_indexed | 2024-09-23T11:47:35Z |
| publishDate | 2008 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/403272019-04-11T14:23:24Z Cutting plane algorithms for variational inference in graphical models Sontag, David Alexander Tommi S. Jaakkola. Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science. Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science. Electrical Engineering and Computer Science. Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2007. This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. Includes bibliographical references (leaves 65-66). In this thesis, we give a new class of outer bounds on the marginal polytope, and propose a cutting-plane algorithm for efficiently optimizing over these constraints. When combined with a concave upper bound on the entropy, this gives a new variational inference algorithm for probabilistic inference in discrete Markov Random Fields (MRFs). Valid constraints are derived for the marginal polytope through a series of projections onto the cut polytope. Projecting onto a larger model gives an efficient separation algorithm for a large class of valid inequalities arising from each of the original projections. As a result, we obtain tighter upper bounds on the logpartition function than possible with previous variational inference algorithms. We also show empirically that our approximations of the marginals are significantly more accurate. This algorithm can also be applied to the problem of finding the Maximum a Posteriori assignment in a MRF, which corresponds to a linear program over the marginal polytope. One of the main contributions of the thesis is to bring together two seemingly different fields, polyhedral combinatorics and probabilistic inference, showing how certain results in either field can carry over to the other. by David Alexander Sontag. S.M. 2008-02-27T20:38:53Z 2008-02-27T20:38:53Z 2007 2007 Thesis http://hdl.handle.net/1721.1/40327 191958182 eng M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission. http://dspace.mit.edu/handle/1721.1/7582 66 leaves application/pdf Massachusetts Institute of Technology |
| spellingShingle | Electrical Engineering and Computer Science. Sontag, David Alexander Cutting plane algorithms for variational inference in graphical models |
| title | Cutting plane algorithms for variational inference in graphical models |
| title_full | Cutting plane algorithms for variational inference in graphical models |
| title_fullStr | Cutting plane algorithms for variational inference in graphical models |
| title_full_unstemmed | Cutting plane algorithms for variational inference in graphical models |
| title_short | Cutting plane algorithms for variational inference in graphical models |
| title_sort | cutting plane algorithms for variational inference in graphical models |
| topic | Electrical Engineering and Computer Science. |
| url | http://hdl.handle.net/1721.1/40327 |
| work_keys_str_mv | AT sontagdavidalexander cuttingplanealgorithmsforvariationalinferenceingraphicalmodels |