ERA

Download the full-sized PDF of Dual Representations for Dynamic ProgrammingDownload the full-sized PDF

Analytics

Share

Permanent link (DOI): https://doi.org/10.7939/R3959C81D

Download

Export to: EndNote  |  Zotero  |  Mendeley

Communities

This file is in the following communities:

Computing Science, Department of

Collections

This file is in the following collections:

Technical Reports (Computing Science)

Dual Representations for Dynamic Programming Open Access

Descriptions

Author or creator
Wang, Tao
Lizotte, Daniel
Bowling, Michael
Schuurmans, Dale
Additional contributors
Subject/Keyword
dynamic programming
Type of item
Computing Science Technical Report
Computing science technical report ID
TR07-10
Language
English
Place
Time
Description
Technical report TR07-10. We propose to use a new dual approach to dynamic programming. The idea is to maintain an explicit representation of stationary distributions as opposed to value functions. A significant advantage of the dual approach is that it allows one to exploit well developed techniques for representing, approximating, and estimating probability distributions, without running the risks associated with divergent value function estimation. In this paper, we introduce novel dual representations and develop dual algorithms. Moreover, we show that dual algorithms remain stable in situations where standard value function estimation diverges. We also show that the dual view yields a viable alternative to standard value function based techniques and opens new avenues for solving sequential decision making problems.
Date created
2007
DOI
doi:10.7939/R3959C81D
License information
Creative Commons Attribution 3.0 Unported
Rights

Citation for previous publication

Source
Link to related item

File Details

Date Uploaded
Date Modified
2014-04-29T20:17:24.466+00:00
Audit Status
Audits have not yet been run on this file.
Characterization
File format: pdf (Portable Document Format)
Mime type: application/pdf
File size: 239252
Last modified: 2015:10:12 21:27:59-06:00
Filename: TR07-10.pdf
Original checksum: c97ee0b465ae3946ed8b3aaf8959e523
Well formed: true
Valid: true
File title: dual.dvi
Page count: 25
Activity of users you follow
User Activity Date