Formal methods for responsibility reasoning in multiagent systems
Formal methods for responsibility reasoning in multiagent systems
Safe and reliable deployment of collaborative AI-human multiagent systems requires formal semantics and verifiable tools to reason about forward-lookingresponsibilities of agents (e.g., who can/should ensure some properties in prospect) aswell as their backward-looking responsibilities (e.g., who to blame, praise, or see accountable for an already materialized outcome in retrospect) [3, 2, 1]. Modeling andreasoning about responsibility calls for capturing its strategic, epistemic, and normativeaspects for which the community of formal methods possesses apt semantic systemsand reasoning tools. In this talk, we report on a line of research on the application offormal methods and modal logics for reasoning about different forms of responsibilityin multiagent systems [4, 5, 6]. In addition, we overview recent work on the application of responsibility reasoning for task coordination, discuss open problems [7], andhighlight the potentials of formal responsibility reasoning in AI systems.
Multiagent Systems, Responsibility Reasoning
Yazdanpanah, Vahid
28f82058-5e51-4f56-be14-191ab5767d56
Dastani, Mehdi
44cecb91-95c6-4821-a307-c43e9434ea4a
15 April 2021
Yazdanpanah, Vahid
28f82058-5e51-4f56-be14-191ab5767d56
Dastani, Mehdi
44cecb91-95c6-4821-a307-c43e9434ea4a
Yazdanpanah, Vahid and Dastani, Mehdi
(2021)
Formal methods for responsibility reasoning in multiagent systems.
Third International Workshop on Formal Methods in Artificial Intelligence (FMAI 2021), Virtual, London, United Kingdom.
15 - 16 Apr 2021.
1 pp
.
Record type:
Conference or Workshop Item
(Other)
Abstract
Safe and reliable deployment of collaborative AI-human multiagent systems requires formal semantics and verifiable tools to reason about forward-lookingresponsibilities of agents (e.g., who can/should ensure some properties in prospect) aswell as their backward-looking responsibilities (e.g., who to blame, praise, or see accountable for an already materialized outcome in retrospect) [3, 2, 1]. Modeling andreasoning about responsibility calls for capturing its strategic, epistemic, and normativeaspects for which the community of formal methods possesses apt semantic systemsand reasoning tools. In this talk, we report on a line of research on the application offormal methods and modal logics for reasoning about different forms of responsibilityin multiagent systems [4, 5, 6]. In addition, we overview recent work on the application of responsibility reasoning for task coordination, discuss open problems [7], andhighlight the potentials of formal responsibility reasoning in AI systems.
Text
FMAI2020_Talk_VY_and_MD
- Author's Original
More information
Published date: 15 April 2021
Venue - Dates:
Third International Workshop on Formal Methods in Artificial Intelligence (FMAI 2021), Virtual, London, United Kingdom, 2021-04-15 - 2021-04-16
Keywords:
Multiagent Systems, Responsibility Reasoning
Identifiers
Local EPrints ID: 449404
URI: http://eprints.soton.ac.uk/id/eprint/449404
PURE UUID: 455e8da5-c338-4734-8fa1-a31658db90e7
Catalogue record
Date deposited: 27 May 2021 16:31
Last modified: 16 Apr 2024 01:58
Export record
Contributors
Author:
Vahid Yazdanpanah
Author:
Mehdi Dastani
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics