Abstract
A decentralized multiagent systems (MAS) is comprised of autonomous agents who interact with each other via asynchronous messaging. A protocol specifies a MAS by specifying the constraints on messaging between agents. Agents enact protocols by applying their own internal decision making. Various kinds of faults may occur when enacting a protocol. For example, messages may be lost, duplicates may be delivered, and agents may crash during the processing of a message. Our contribution in this paper is demonstrating how information protocols support rich fault tolerance mechanisms, and in a manner that is unanticipated by alternative approaches for engineering decentralized MAS.
Cite
CITATION STYLE
Christie V, S. H., & Chopra, A. K. (2020). Fault Tolerance in Multiagent Systems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 12589 LNAI, pp. 78–86). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-66534-0_5
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
