Addressing trust and mutability issues in XAI utilising case based reasoning.

Salimi, Pedram

Abstract

Explainable AI (XAI) research is required to ensure that explanations are human readable and understandable. The present XAI approaches are useful for observing and comprehending some of the most important underlying properties of any Black-box AI model. However, when it comes to pushing them into production, certain critical concerns may arise: (1) How can end-users rely on the output of an XAI platform and trust the system? (2) How can end-users customise the platform's output depending on their own preferences In this project, we will explore how to address these concerns by utilising Cased-based Reasoning. Accordingly, we propose to exploit the neighbourhood to improve end-user trust by offering similar cases and confidence scores and using different retrieval strategies to address end-user preferences. Additionally, this project will also look at how to leverage Conversational AI and Natural Language Generation approaches to improve the interactive and engaging user experience with example-based XAI systems.

Citation

SALIMI, P. 2022. Addressing trust and mutability issues in XAI utilising case based reasoning. In Reuss, P. and Schönborn, J. (eds.) Proceedings of the 30th Doctoral consortium of the international conference on case-based reasoning (ICCBR-DC 2022), co-located with the 30th International conference on case-based reasoning (ICCBR 2022), 12-15 September 2022, Nancy, France. CEUR workshop proceedings, 3418. Aachen: CEUR-WS [online], pages 22-27. Available from: https://ceur-ws.org/Vol-3418/ICCBR_2022_DC_paper19.pdf

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 30th Doctoral consortium of the international conference on case-based reasoning (ICCBR-DC 2022), co-located with the 30th International conference on case-based reasoning (ICCBR 2022)
• Start Date: Sep 12, 2022
• End Date: Sep 15, 2022
• Acceptance Date: Jul 22, 2022
• Online Publication Date: Sep 15, 2022
• Publication Date: Jun 13, 2023
• Deposit Date: Jul 21, 2023
• Publicly Available Date: Jul 21, 2023
• Publisher: CEUR-WS
• Peer Reviewed: Peer Reviewed
• Pages: 22-27
• Series Title: CEUR workshop proceedings
• Series Number: 3418
• Series ISSN: 1613-0073
• Keywords: Explainable AI; Cased-based reasoning; Conversational AI; Natural language generation
• Public URL: https://rgu-repository.worktribe.com/output/2015560
• Publisher URL: https://ceur-ws.org/Vol-3418/ICCBR_2022_DC_paper19.pdf