Modified policy iteration algorithms are not strongly polynomial for discounted dynamic programming
Year of publication: |
2014
|
---|---|
Authors: | Feinberg, Eugene A. ; Huang, Jefferson ; Scherrer, Bruno |
Published in: |
Operations research letters. - Amsterdam [u.a.] : Elsevier, ISSN 0167-6377, ZDB-ID 720735-9. - Vol. 42.2014, 6/7, p. 429-431
|
Subject: | Markov decision process | Modified policy iteration | Strongly polynomial | Policy | Algorithm | Mathematische Optimierung | Mathematical programming | Theorie | Theory | Markov-Kette | Markov chain | Algorithmus |
-
The value iteration algorithm is not strongly polynomial for discounted dynamic programming
Feinberg, Eugene A., (2014)
-
Wang, Mengdi, (2020)
-
Complexity bounds for approximately solving discounted MDPs by value iterations
Feinberg, Eugene A., (2020)
- More ...
-
Feinberg, Eugene A., (2013)
-
The value iteration algorithm is not strongly polynomial for discounted dynamic programming
Feinberg, Eugene A., (2014)
-
Feinberg, Eugene A., (2018)
- More ...