2014
DOI: 10.1007/978-3-319-11508-5_15
|View full text |Cite
|
Sign up to set email alerts
|

Solving Hidden-Semi-Markov-Mode Markov Decision Problems

Abstract: Abstract. Hidden-Mode Markov Decision Processes (HM-MDPs) were proposed to represent sequential decision-making problems in non-stationary environments that evolve according to a Markov chain. We introduce in this paper Hidden-Semi-Markov-Mode Markov Decision Processes (HS3MDPs), a generalization of HM-MDPs to the more realistic case of non-stationary environments evolving according to a semi-Markov chain. Like HM-MDPs, HS3MDPs form a subclass of Partially Observable Markov Decision Processes. Therefore, large… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
10
0

Year Published

2016
2016
2018
2018

Publication Types

Select...
3
3
1

Relationship

2
5

Authors

Journals

citations
Cited by 10 publications
(10 citation statements)
references
References 11 publications
0
10
0
Order By: Relevance
“…Future work will explore more sophisticated models of the adversaries. We would like to include the temporal dimension in the context changes as it could have been done in HS3MDPs 28 . Models inspired from behavioral economics could also be useful to develop more accurate profile of the adversaries.…”
Section: Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…Future work will explore more sophisticated models of the adversaries. We would like to include the temporal dimension in the context changes as it could have been done in HS3MDPs 28 . Models inspired from behavioral economics could also be useful to develop more accurate profile of the adversaries.…”
Section: Resultsmentioning
confidence: 99%
“…Following previous works dealing with non-stationary mono-agent POMDPs 28 , we decompose the non-stationary decision problem as a series of stationary decision problems. Each stationary phase is then referred to as a mode or a context.…”
Section: Dec-pomdp Model For a Current Contextmentioning
confidence: 99%
“…Besides, the parameter may not be observable. In that case, a model like Hidden-Semi-Markov-Mode MDP proposed by [8] could be exploited.…”
Section: Resultsmentioning
confidence: 99%
“…Inspired by the notion of modes used in mono-agent POMDP [18], our approach consists in defining a DEC-POMDP for a given distribution P I which will be referred as the current context of the decision making. As discussed later in the paper, the DEC-POMDP definition will have to be updated as the context evolves.…”
Section: B Patrollers' Dec-pomdp Formulationmentioning
confidence: 99%