Planning under Uncertainty with Weighted State Scenarios

Erwin Walraven; Matthijs T. J. Spaan

Planning under Uncertainty with Weighted State Scenarios

Algorithmics

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

3 Downloads (Pure)

Abstract

External factors are hard to model using a Markovian state in several real-world planning domains. Although planning can be difficult in such domains, it may be possible to exploit long-term dependencies between states of the environment during planning. We introduce weighted state scenarios to model long-term sequences of states, and we use a model based on a Partially Observable Markov Decision Process to reason about scenarios during planning. Experiments show that our model outperforms other methods for decision making in two real-world domains.

Original language	English
Title of host publication	AAAI 2015 Fall Symposium on Sequential Decision Making for Intelligent Agents
Publisher	American Association for Artificial Intelligence (AAAI)
Pages	93-94
Number of pages	2
Publication status	Published - 2015
Event	AAAI 2015 Fall Symposium Sequential Decision Making for Intelligent Agents - Arlington, Virginia, United States Duration: 12 Nov 2015 → 14 Nov 2015 http://masplan.org/sdmia

Conference

Conference	AAAI 2015 Fall Symposium Sequential Decision Making for Intelligent Agents
Country/Territory	United States
City	Arlington, Virginia
Period	12/11/15 → 14/11/15
Internet address	http://masplan.org/sdmia

Bibliographical note

Extended abstract

Keywords

Markov decision processes
planning under uncertainty
renewable energy
smart grids

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

http://www.aaai.org/ocs/index.php/FSS/FSS15/paper/view/11689/11514

GCP: Gaming beyond the Copper Plate
de Weerdt, M. M., Spaan, M. T. J., de Vries, L. J., Witteveen, C., van der Sluis, L., Walraven, E. M. P., Philipsen, R. M., Morales Espana, G. A. & Ramirez Elizondo, L. M.
1/10/14 → 31/10/18
Project: Research

Cite this

@inproceedings{8499d63ac7fc42149f1830ababfaa62a,

title = "Planning under Uncertainty with Weighted State Scenarios",

abstract = "External factors are hard to model using a Markovian state in several real-world planning domains. Although planning can be difficult in such domains, it may be possible to exploit long-term dependencies between states of the environment during planning. We introduce weighted state scenarios to model long-term sequences of states, and we use a model based on a Partially Observable Markov Decision Process to reason about scenarios during planning. Experiments show that our model outperforms other methods for decision making in two real-world domains.",

keywords = "Markov decision processes, planning under uncertainty, renewable energy, smart grids",

author = "Erwin Walraven and Spaan, {Matthijs T. J.}",

note = "Extended abstract; AAAI 2015 Fall Symposium Sequential Decision Making for Intelligent Agents ; Conference date: 12-11-2015 Through 14-11-2015",

year = "2015",

language = "English",

pages = "93--94",

booktitle = "AAAI 2015 Fall Symposium on Sequential Decision Making for Intelligent Agents",

publisher = "American Association for Artificial Intelligence (AAAI)",

address = "United States",

url = "http://masplan.org/sdmia",

}

Walraven, E & Spaan, MTJ 2015, Planning under Uncertainty with Weighted State Scenarios. in AAAI 2015 Fall Symposium on Sequential Decision Making for Intelligent Agents. American Association for Artificial Intelligence (AAAI), pp. 93-94, AAAI 2015 Fall Symposium Sequential Decision Making for Intelligent Agents, Arlington, Virginia, United States, 12/11/15. <http://www.aaai.org/ocs/index.php/FSS/FSS15/paper/view/11689/11514>

Planning under Uncertainty with Weighted State Scenarios. / Walraven, Erwin; Spaan, Matthijs T. J.
AAAI 2015 Fall Symposium on Sequential Decision Making for Intelligent Agents. American Association for Artificial Intelligence (AAAI), 2015. p. 93-94.

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Planning under Uncertainty with Weighted State Scenarios

AU - Walraven, Erwin

AU - Spaan, Matthijs T. J.

N1 - Extended abstract

PY - 2015

Y1 - 2015

N2 - External factors are hard to model using a Markovian state in several real-world planning domains. Although planning can be difficult in such domains, it may be possible to exploit long-term dependencies between states of the environment during planning. We introduce weighted state scenarios to model long-term sequences of states, and we use a model based on a Partially Observable Markov Decision Process to reason about scenarios during planning. Experiments show that our model outperforms other methods for decision making in two real-world domains.

AB - External factors are hard to model using a Markovian state in several real-world planning domains. Although planning can be difficult in such domains, it may be possible to exploit long-term dependencies between states of the environment during planning. We introduce weighted state scenarios to model long-term sequences of states, and we use a model based on a Partially Observable Markov Decision Process to reason about scenarios during planning. Experiments show that our model outperforms other methods for decision making in two real-world domains.

KW - Markov decision processes

KW - planning under uncertainty

KW - renewable energy

KW - smart grids

M3 - Conference contribution

SP - 93

EP - 94

BT - AAAI 2015 Fall Symposium on Sequential Decision Making for Intelligent Agents

PB - American Association for Artificial Intelligence (AAAI)

T2 - AAAI 2015 Fall Symposium Sequential Decision Making for Intelligent Agents

Y2 - 12 November 2015 through 14 November 2015

ER -

Planning under Uncertainty with Weighted State Scenarios

Abstract

Conference

Bibliographical note

Keywords

UN SDGs

Access to Document

Fingerprint

Projects

GCP: Gaming beyond the Copper Plate

Cite this