IMDEA Networks Institute Publications Repository

Human-driven Decision-making in Accessing Distributed Resources

Stavrakakis, Ioannis (2017) Human-driven Decision-making in Accessing Distributed Resources. In: Distinguished Lecture Series, The Collaborative Research Centre (CRC) 1053 "MAKI – Multi-Mechanisms Adaptation for the Future Internet", Technische Universität Darmstadt, 29 June 2017, Darmstadt, Germany.

Full text not available from this repository.

Abstract

Information and Communication Technologies enable the generation and dissemination of information that can enhance tremendously our awareness about the environment and its resources. This resource awareness provides enhanced service opportunities but it may also create contention and congestion penalties in distributed, uncoordinated resource environments, to the point of making such information provision counter-productive. This talk employs a simple binary model that can represent a wide range of decision formulations for this environment (to compete or not to compete for the resources) with a ternary cost structure (success vs failure vs not competing). The case of fully rational (strategic) decision-making is briefly presented and the cost of anarchy (due to the lack of coordination) is discussed. Since humans who ultimately take decisions have limitations, some models for human-driven decision making are presented, reflecting computational/cognitive human limitations or biases. It is shown that bounded rationality decisions can be as or even more effective than those under the full rationality model. Finally, and as time permits, alternative to the fully uncoordinated approaches are discussed in an effort to reduce the cost of anarchy, along with their potential benefits and drawbacks.

Item Type: Conference or Workshop Papers (Invited Talk)
Subjects: UNSPECIFIED
Divisions: UNSPECIFIED
Depositing User: Rebeca De Miguel
Date Deposited: 21 Sep 2017 10:09
Last Modified: 21 Sep 2017 10:09
URI: http://eprints.networks.imdea.org/id/eprint/1665

Actions (login required)

View Item View Item