Adaptive Envelope MDPs for Relational Equivalence-based Planning
We describe a method to use structured representations of the environmentâs dynamics to constrain and speed up the planning process. Given a problem domain described in a probabilistic logical description language, we develop an anytime technique that incrementally improves on an initial, partial...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Published: |
2008
|
| Online Access: | http://hdl.handle.net/1721.1/41920 |
| _version_ | 1826206346729488384 |
|---|---|
| author | Gardiol, Natalia H. Kaelbling, Leslie Pack |
| author2 | Leslie Kaelbling |
| author_facet | Leslie Kaelbling Gardiol, Natalia H. Kaelbling, Leslie Pack |
| author_sort | Gardiol, Natalia H. |
| collection | MIT |
| description | We describe a method to use structured representations of the environmentâs dynamics to constrain and speed up the planning process. Given a problem domain described in a probabilistic logical description language, we develop an anytime technique that incrementally improves on an initial, partial policy. This partial solution is found by ï¬rst reducing the number of predicates needed to represent a relaxed version of the problem to a minimum, and then dynamically partitioning the action space into a set of equivalence classes with respect to this minimal representation. Our approach uses the envelope MDP framework, which creates a Markov decision process out of a subset of the full state space as de- termined by the initial partial solution. This strategy permits an agent to begin acting within a restricted part of the full state space and to expand its envelope judiciously as resources permit. |
| first_indexed | 2024-09-23T13:27:45Z |
| id | mit-1721.1/41920 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T13:27:45Z |
| publishDate | 2008 |
| record_format | dspace |
| spelling | mit-1721.1/419202019-04-12T09:43:45Z Adaptive Envelope MDPs for Relational Equivalence-based Planning Gardiol, Natalia H. Kaelbling, Leslie Pack Leslie Kaelbling Learning and Intelligent Systems We describe a method to use structured representations of the environmentâs dynamics to constrain and speed up the planning process. Given a problem domain described in a probabilistic logical description language, we develop an anytime technique that incrementally improves on an initial, partial policy. This partial solution is found by ï¬rst reducing the number of predicates needed to represent a relaxed version of the problem to a minimum, and then dynamically partitioning the action space into a set of equivalence classes with respect to this minimal representation. Our approach uses the envelope MDP framework, which creates a Markov decision process out of a subset of the full state space as de- termined by the initial partial solution. This strategy permits an agent to begin acting within a restricted part of the full state space and to expand its envelope judiciously as resources permit. 2008-08-01T21:30:16Z 2008-08-01T21:30:16Z 2008-07-29 MIT-CSAIL-TR-2008-050 http://hdl.handle.net/1721.1/41920 Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory 17 p. application/pdf application/postscript |
| spellingShingle | Gardiol, Natalia H. Kaelbling, Leslie Pack Adaptive Envelope MDPs for Relational Equivalence-based Planning |
| title | Adaptive Envelope MDPs for Relational Equivalence-based Planning |
| title_full | Adaptive Envelope MDPs for Relational Equivalence-based Planning |
| title_fullStr | Adaptive Envelope MDPs for Relational Equivalence-based Planning |
| title_full_unstemmed | Adaptive Envelope MDPs for Relational Equivalence-based Planning |
| title_short | Adaptive Envelope MDPs for Relational Equivalence-based Planning |
| title_sort | adaptive envelope mdps for relational equivalence based planning |
| url | http://hdl.handle.net/1721.1/41920 |
| work_keys_str_mv | AT gardiolnataliah adaptiveenvelopemdpsforrelationalequivalencebasedplanning AT kaelblinglesliepack adaptiveenvelopemdpsforrelationalequivalencebasedplanning |