Adapting Reinforcement Learning For Trust: Effective Modeling in Dynamic Environments (Short Paper). / Kafali, Ozgur; Yolum, Pinar.

IEEE / WIC / ACM International Conference on Web Intelligence. 2009. p. 383-386.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Published

Standard

Adapting Reinforcement Learning For Trust: Effective Modeling in Dynamic Environments (Short Paper). / Kafali, Ozgur; Yolum, Pinar.

IEEE / WIC / ACM International Conference on Web Intelligence. 2009. p. 383-386.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Harvard

Kafali, O & Yolum, P 2009, Adapting Reinforcement Learning For Trust: Effective Modeling in Dynamic Environments (Short Paper). in IEEE / WIC / ACM International Conference on Web Intelligence. pp. 383-386. <http://dl.acm.org/citation.cfm?id=1632360>

APA

Kafali, O., & Yolum, P. (2009). Adapting Reinforcement Learning For Trust: Effective Modeling in Dynamic Environments (Short Paper). In IEEE / WIC / ACM International Conference on Web Intelligence (pp. 383-386) http://dl.acm.org/citation.cfm?id=1632360

Vancouver

Kafali O, Yolum P. Adapting Reinforcement Learning For Trust: Effective Modeling in Dynamic Environments (Short Paper). In IEEE / WIC / ACM International Conference on Web Intelligence. 2009. p. 383-386

Author

Kafali, Ozgur ; Yolum, Pinar. / Adapting Reinforcement Learning For Trust: Effective Modeling in Dynamic Environments (Short Paper). IEEE / WIC / ACM International Conference on Web Intelligence. 2009. pp. 383-386

BibTeX

@inproceedings{eae1d4c0c27b4bcaafa4cc9b5857dad6,
title = "Adapting Reinforcement Learning For Trust: Effective Modeling in Dynamic Environments (Short Paper)",
abstract = "In open multiagent systems, agents need to model their environments in order to identify trustworthy agents. Models of the environment should be accurate so that decisions about whom to interact with can be done soundly. Traditional trust models are based on modeling specific properties of agents, such as their expertise or reliability. Building those models requires too many prior interactions to be accurate. This paper proposes an approach that is based on keeping track of outcomes of agent's actions towards others rather than modeling other agents' performances explicitly. Contrary to existing modeling approaches that require domain knowledge to build models, our proposed approach can be effectively realized in multiagent systems when the agent's actions are clearly identified. Comparisons with other modeling approaches in various environments reveal that our proposed approach can create more precise models in short time and can adjust its behavior quickly when other agents' behaviors change.",
author = "Ozgur Kafali and Pinar Yolum",
year = "2009",
language = "English",
pages = "383--386",
booktitle = "IEEE / WIC / ACM International Conference on Web Intelligence",

}

RIS

TY - GEN

T1 - Adapting Reinforcement Learning For Trust: Effective Modeling in Dynamic Environments (Short Paper)

AU - Kafali, Ozgur

AU - Yolum, Pinar

PY - 2009

Y1 - 2009

N2 - In open multiagent systems, agents need to model their environments in order to identify trustworthy agents. Models of the environment should be accurate so that decisions about whom to interact with can be done soundly. Traditional trust models are based on modeling specific properties of agents, such as their expertise or reliability. Building those models requires too many prior interactions to be accurate. This paper proposes an approach that is based on keeping track of outcomes of agent's actions towards others rather than modeling other agents' performances explicitly. Contrary to existing modeling approaches that require domain knowledge to build models, our proposed approach can be effectively realized in multiagent systems when the agent's actions are clearly identified. Comparisons with other modeling approaches in various environments reveal that our proposed approach can create more precise models in short time and can adjust its behavior quickly when other agents' behaviors change.

AB - In open multiagent systems, agents need to model their environments in order to identify trustworthy agents. Models of the environment should be accurate so that decisions about whom to interact with can be done soundly. Traditional trust models are based on modeling specific properties of agents, such as their expertise or reliability. Building those models requires too many prior interactions to be accurate. This paper proposes an approach that is based on keeping track of outcomes of agent's actions towards others rather than modeling other agents' performances explicitly. Contrary to existing modeling approaches that require domain knowledge to build models, our proposed approach can be effectively realized in multiagent systems when the agent's actions are clearly identified. Comparisons with other modeling approaches in various environments reveal that our proposed approach can create more precise models in short time and can adjust its behavior quickly when other agents' behaviors change.

M3 - Conference contribution

SP - 383

EP - 386

BT - IEEE / WIC / ACM International Conference on Web Intelligence

ER -