Explainability of AI-driven air combat agent

Simple item page
dc.contributor.authorSaldiran, Emre
dc.contributor.authorHasanzade, Mehmet
dc.contributor.authorInalhan, Gokhan
dc.contributor.authorTsourdos, Antonios
dc.date.accessioned2023-09-15T13:41:28Z
dc.date.available2023-09-15T13:41:28Z
dc.date.issued2023-08-02
dc.description.abstractIn safety-critical applications, it is crucial to verify and certify the decisions made by AI-driven Autonomous Systems (ASs). However, the black-box nature of neural networks used in these systems often makes it challenging to achieve this. The explainability of these systems can help with the verification and certification process, which will speed up their deployment in safety-critical applications. This study investigates the explainability of AI-driven air combat agents via semantically grouped reward decomposition. The paper presents two use cases to demonstrate how this approach can help AI and non-AI experts to evaluate and debug the behavior of RL agents.en_UK
dc.description.sponsorshipBAE Systemsen_UK
dc.identifier.citationSaldiran E, Hasanzade M, Inalhan G, Tsourdos A. (2023) Explainability of AI-driven air combat agent. In: 2023 IEEE Conference on Artificial Intelligence (CAI 2023), 5-6 June 2023, Santa Clara, USA, pp. 85-86en_UK
dc.identifier.eisbn979-8-3503-3984-0
dc.identifier.isbn979-8-3503-3985-7
dc.identifier.urihttps://doi.org/10.1109/CAI54212.2023.00044
dc.identifier.urihttps://dspace.lib.cranfield.ac.uk/handle/1826/20220
dc.language.isoenen_UK
dc.publisherIEEEen_UK
dc.rightsAttribution-NonCommercial 4.0 International*
dc.rights.urihttp://creativecommons.org/licenses/by-nc/4.0/*
dc.subjectexplainableen_UK
dc.subjectreinforcement learningen_UK
dc.subjectreward decompositionen_UK
dc.subjectair combaten_UK
dc.titleExplainability of AI-driven air combat agenten_UK
dc.typeConference paperen_UK

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
AI-driven_air_combat_agent-2023.pdf
Size:
3.05 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.63 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Staff publications (SATM)