iSee: Advancing Multi-Shot Explainable AI Using Case-based Recommendations

TL;DR

The paper addresses the need for multi-shot explainable AI by proposing iSee, a platform that harnesses Case-based Reasoning to share and reuse explanation experiences. It formalizes explanations as Explanation Experience Cases within a CBR 4R cycle and uses a Behavior Tree-based explanation strategy, along with a microservice architecture and an interoperable iSee ontology. A radiograph classification use case demonstrates retrieval, adaptation, revision, and retention of explanation strategies. A summative mixed-methods evaluation with six design users shows that iSee generalizes across applications and supports adoption of XAI best practices, while highlighting usability strengths and security concerns to address in future work.

Abstract

Explainable AI (XAI) can greatly enhance user trust and satisfaction in AI-assisted decision-making processes. Recent findings suggest that a single explainer may not meet the diverse needs of multiple users in an AI system; indeed, even individual users may require multiple explanations. This highlights the necessity for a "multi-shot" approach, employing a combination of explainers to form what we introduce as an "explanation strategy". Tailored to a specific user or a user group, an "explanation experience" describes interactions with personalised strategies designed to enhance their AI decision-making processes. The iSee platform is designed for the intelligent sharing and reuse of explanation experiences, using Case-based Reasoning to advance best practices in XAI. The platform provides tools that enable AI system designers, i.e. design users, to design and iteratively revise the most suitable explanation strategy for their AI system to satisfy end-user needs. All knowledge generated within the iSee platform is formalised by the iSee ontology for interoperability. We use a summative mixed methods study protocol to evaluate the usability and utility of the iSee platform with six design users across varying levels of AI and XAI expertise. Our findings confirm that the iSee platform effectively generalises across applications and its potential to promote the adoption of XAI best practices.

Figures (8)

• Figure 1: Explanation Experience Case
• Figure 2: Sample Case Solution
• Figure 3: Failure Driven Adaptation
• Figure 5: Radiograph Classification Use Case created following the iSee CBR 4Rs - Please refer to the supplementary material to access the full-resolution screenshots of the platform
• Figure 6: iSee Cockpit time taken per component