Maximizing expected impact in an agent reputation network

Gavin Rens; Abhaya Nayak; Thomas Meyer

doi:10.1007/978-3-030-00111-7_9

Maximizing expected impact in an agent reputation network

Gavin Rens^*, Abhaya Nayak, Thomas Meyer

^*Corresponding author for this work

Department of Computing

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution

Abstract

We propose a new framework for reasoning about the reputation of multiple agents, based on the partially observable Markov decision process (POMDP). It is general enough for the specification of a variety of stochastic multi-agent system (MAS) domains involving the impact of agents on each other’s reputations. Assuming that an agent must maintain a good enough reputation to survive in the system, a method for an agent to select optimal actions is developed.

Original language	English
Title of host publication	KI 2018
Subtitle of host publication	Advances in Artificial Intelligence - 41st German Conference on AI, 2018, Proceedings
Editors	Frank Trollmann, Anni-Yasmin Turhan
Place of Publication	Switzerland
Publisher	Springer-VDI-Verlag GmbH & Co. KG
Pages	99-106
Number of pages	8
ISBN (Electronic)	9783030001117
ISBN (Print)	9783030001100
DOIs	https://doi.org/10.1007/978-3-030-00111-7_9
Publication status	Published - 1 Jan 2018
Event	41st German Conference on Artificial Intelligence, KI 2018 - Berlin, Germany Duration: 24 Sep 2018 → 28 Sep 2018

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11117 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	41st German Conference on Artificial Intelligence, KI 2018
Country	Germany
City	Berlin
Period	24/09/18 → 28/09/18

Keywords

Planning
POMDP
Trust and reputation
Uncertainty

Access to Document

10.1007/978-3-030-00111-7_9

Link to publication in Scopus

Fingerprint Dive into the research topics of 'Maximizing expected impact in an agent reputation network'. Together they form a unique fingerprint.

Cite this

Rens, G., Nayak, A., & Meyer, T. (2018). Maximizing expected impact in an agent reputation network. In F. Trollmann, & A-Y. Turhan (Eds.), KI 2018: Advances in Artificial Intelligence - 41st German Conference on AI, 2018, Proceedings (pp. 99-106). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11117 LNAI). Switzerland: Springer-VDI-Verlag GmbH & Co. KG. https://doi.org/10.1007/978-3-030-00111-7_9

Rens, Gavin ; Nayak, Abhaya ; Meyer, Thomas. / Maximizing expected impact in an agent reputation network. KI 2018: Advances in Artificial Intelligence - 41st German Conference on AI, 2018, Proceedings. editor / Frank Trollmann ; Anni-Yasmin Turhan. Switzerland : Springer-VDI-Verlag GmbH & Co. KG, 2018. pp. 99-106 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{62be8423a41c42098b8c0b5abc0c255b,

title = "Maximizing expected impact in an agent reputation network",

abstract = "We propose a new framework for reasoning about the reputation of multiple agents, based on the partially observable Markov decision process (POMDP). It is general enough for the specification of a variety of stochastic multi-agent system (MAS) domains involving the impact of agents on each other’s reputations. Assuming that an agent must maintain a good enough reputation to survive in the system, a method for an agent to select optimal actions is developed.",

keywords = "Planning, POMDP, Trust and reputation, Uncertainty",

author = "Gavin Rens and Abhaya Nayak and Thomas Meyer",

year = "2018",

month = "1",

day = "1",

doi = "10.1007/978-3-030-00111-7_9",

language = "English",

isbn = "9783030001100",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer-VDI-Verlag GmbH & Co. KG",

pages = "99--106",

editor = "Frank Trollmann and Anni-Yasmin Turhan",

booktitle = "KI 2018",

address = "Germany",

}

Rens, G, Nayak, A & Meyer, T 2018, Maximizing expected impact in an agent reputation network. in F Trollmann & A-Y Turhan (eds), KI 2018: Advances in Artificial Intelligence - 41st German Conference on AI, 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11117 LNAI, Springer-VDI-Verlag GmbH & Co. KG, Switzerland, pp. 99-106, 41st German Conference on Artificial Intelligence, KI 2018, Berlin, Germany, 24/09/18. https://doi.org/10.1007/978-3-030-00111-7_9

Maximizing expected impact in an agent reputation network. / Rens, Gavin; Nayak, Abhaya; Meyer, Thomas.

KI 2018: Advances in Artificial Intelligence - 41st German Conference on AI, 2018, Proceedings. ed. / Frank Trollmann; Anni-Yasmin Turhan. Switzerland : Springer-VDI-Verlag GmbH & Co. KG, 2018. p. 99-106 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11117 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution

TY - GEN

T1 - Maximizing expected impact in an agent reputation network

AU - Rens, Gavin

AU - Nayak, Abhaya

AU - Meyer, Thomas

PY - 2018/1/1

Y1 - 2018/1/1

N2 - We propose a new framework for reasoning about the reputation of multiple agents, based on the partially observable Markov decision process (POMDP). It is general enough for the specification of a variety of stochastic multi-agent system (MAS) domains involving the impact of agents on each other’s reputations. Assuming that an agent must maintain a good enough reputation to survive in the system, a method for an agent to select optimal actions is developed.

AB - We propose a new framework for reasoning about the reputation of multiple agents, based on the partially observable Markov decision process (POMDP). It is general enough for the specification of a variety of stochastic multi-agent system (MAS) domains involving the impact of agents on each other’s reputations. Assuming that an agent must maintain a good enough reputation to survive in the system, a method for an agent to select optimal actions is developed.

KW - Planning

KW - POMDP

KW - Trust and reputation

KW - Uncertainty

UR - http://www.scopus.com/inward/record.url?scp=85054469861&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-00111-7_9

DO - 10.1007/978-3-030-00111-7_9

M3 - Conference proceeding contribution

SN - 9783030001100

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 99

EP - 106

BT - KI 2018

A2 - Trollmann, Frank

A2 - Turhan, Anni-Yasmin

PB - Springer-VDI-Verlag GmbH & Co. KG

CY - Switzerland

ER -

Rens G, Nayak A, Meyer T. Maximizing expected impact in an agent reputation network. In Trollmann F, Turhan A-Y, editors, KI 2018: Advances in Artificial Intelligence - 41st German Conference on AI, 2018, Proceedings. Switzerland: Springer-VDI-Verlag GmbH & Co. KG. 2018. p. 99-106. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-00111-7_9