Teaching and leading an ad hoc teammate: Collaboration without pre-coordination

Peter Stone; Gal A. Kaminka; Sarit Kraus; Jeffrey S. Rosenschein; Noa Agmon

doi:10.1016/j.artint.2013.07.003

Teaching and leading an ad hoc teammate: Collaboration without pre-coordination

Peter Stone, Gal A. Kaminka, Sarit Kraus, Jeffrey S. Rosenschein, Noa Agmon

Department of Computer Science

Research output: Contribution to journal › Article › peer-review

24 Scopus citations

Abstract

As autonomous agents proliferate in the real world, both in software and robotic settings, they will increasingly need to band together for cooperative activities with previously unfamiliar teammates. In such ad hoc team settings, team strategies cannot be developed a priori. Rather, an agent must be prepared to cooperate with many types of teammates: it must collaborate without pre-coordination. This article defines two aspects of collaboration in two-player teams, involving either simultaneous or sequential decision making. In both cases, the ad hoc agent is more knowledgeable of the environment, and attempts to influence the behavior of its teammate such that they will attain the optimal possible joint utility.

Original language	English
Pages (from-to)	35-65
Number of pages	31
Journal	Artificial Intelligence
Volume	203
DOIs	https://doi.org/10.1016/j.artint.2013.07.003
State	Published - 2013

Bibliographical note

Funding Information:
Thanks to Michael Littman and Jeremy Stober for helpful comments pertaining to Section 2 . Thanks to Yonatan Aumann, Vincent Conitzer, Reshef Meir, Daniel Stronger, and Leonid Trainer for helpful comments pertaining to Section 3 . Thanks also to the UT Austin Learning Agents Research Group (LARG) for useful comments and suggestions. This work was partially supported by grants from NSF ( IIS-0917122 , IIS-0705587 ), DARPA ( FA8650-08-C-7812 ), ONR ( N00014-09-1-0658 ), FHWA ( DTFH61-07-H-00030 ), Army Research Lab ( W911NF-08-1-0144 ), ISF ( 1357/07 , 898/05 ), Israel Ministry of Science and Technology ( 3-6797 ), ERC (# 267523 ), MURI ( W911NF-08-1-0144 ) and the Fulbright and Guggenheim Foundations .

Keywords

Autonomous agents
Game theory
Multiagent systems
Teamwork
k-armed bandits

Access to Document

10.1016/j.artint.2013.07.003

Cite this

@article{95c19677ef454ceea261f7202b99ef00,

title = "Teaching and leading an ad hoc teammate: Collaboration without pre-coordination",

abstract = "As autonomous agents proliferate in the real world, both in software and robotic settings, they will increasingly need to band together for cooperative activities with previously unfamiliar teammates. In such ad hoc team settings, team strategies cannot be developed a priori. Rather, an agent must be prepared to cooperate with many types of teammates: it must collaborate without pre-coordination. This article defines two aspects of collaboration in two-player teams, involving either simultaneous or sequential decision making. In both cases, the ad hoc agent is more knowledgeable of the environment, and attempts to influence the behavior of its teammate such that they will attain the optimal possible joint utility.",

keywords = "Autonomous agents, Game theory, Multiagent systems, Teamwork, k-armed bandits",

author = "Peter Stone and Kaminka, {Gal A.} and Sarit Kraus and Rosenschein, {Jeffrey S.} and Noa Agmon",

year = "2013",

doi = "10.1016/j.artint.2013.07.003",

language = "אנגלית",

volume = "203",

pages = "35--65",

journal = "Artificial Intelligence",

issn = "0004-3702",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Teaching and leading an ad hoc teammate

T2 - Collaboration without pre-coordination

AU - Stone, Peter

AU - Kaminka, Gal A.

AU - Kraus, Sarit

AU - Rosenschein, Jeffrey S.

AU - Agmon, Noa

PY - 2013

Y1 - 2013

N2 - As autonomous agents proliferate in the real world, both in software and robotic settings, they will increasingly need to band together for cooperative activities with previously unfamiliar teammates. In such ad hoc team settings, team strategies cannot be developed a priori. Rather, an agent must be prepared to cooperate with many types of teammates: it must collaborate without pre-coordination. This article defines two aspects of collaboration in two-player teams, involving either simultaneous or sequential decision making. In both cases, the ad hoc agent is more knowledgeable of the environment, and attempts to influence the behavior of its teammate such that they will attain the optimal possible joint utility.

AB - As autonomous agents proliferate in the real world, both in software and robotic settings, they will increasingly need to band together for cooperative activities with previously unfamiliar teammates. In such ad hoc team settings, team strategies cannot be developed a priori. Rather, an agent must be prepared to cooperate with many types of teammates: it must collaborate without pre-coordination. This article defines two aspects of collaboration in two-player teams, involving either simultaneous or sequential decision making. In both cases, the ad hoc agent is more knowledgeable of the environment, and attempts to influence the behavior of its teammate such that they will attain the optimal possible joint utility.

KW - Autonomous agents

KW - Game theory

KW - Multiagent systems

KW - Teamwork

KW - k-armed bandits

UR - http://www.scopus.com/inward/record.url?scp=84882311723&partnerID=8YFLogxK

U2 - 10.1016/j.artint.2013.07.003

DO - 10.1016/j.artint.2013.07.003

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

SN - 0004-3702

VL - 203

SP - 35

EP - 65

JO - Artificial Intelligence

JF - Artificial Intelligence

ER -

Teaching and leading an ad hoc teammate: Collaboration without pre-coordination

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this