Operationalizing Human-Centered Perspectives in Explainable AI

Abstract

The realm of Artifcial Intelligence (AI)'s impact on our lives is far reaching -with AI systems proliferating high-stakes domains such as healthcare, fnance, mobility, law, etc., these systems must be able to explain their decision to diverse end-users comprehensibly. Yet the discourse of Explainable AI (XAI) has been predominantly focused on algorithm-centered approaches, sufering from gaps in meeting user needs and exacerbating issues of algorithmic opacity. To address these issues, researchers have called for human-centered approaches to XAI. There is a need to chart the domain and shape the discourse of XAI with refective discussions from diverse stakeholders. The goal of this workshop is to examine how human-centered perspectives in XAI can be operationalized at the conceptual, methodological, and technical levels. Encouraging holistic (historical, sociological, and technical) approaches, we put an emphasis on "operationalizing", aiming to produce actionable frameworks, transferable evaluation methods, concrete design guidelines, and articulate a coordinated research agenda for XAI. CCS CONCEPTS• Human-centered computing → HCI theory, concepts and models; Visualization theory, concepts and paradigms; Visualization design and evaluation methods; • Computing methodologies → Philosophical/theoretical foundations of artifcial intelligence.

Investigating Explainability of Generative AI for Code through Scenario-based Design

“…A full review of XAI techniques is beyond the scope of this paper and can be found in many recent survey papers [1,29,55,56]. Our work is most closely informed by, and intends to bridge, the emerging topic of explainability for generative models, and the inter-disciplinary field of Human-Centered Explainable AI (HCXAI) [21,22,52].…”

“…The term Human-centered AI has emerged in many academic works and public discussions [22,25,45,65,85,89]. While definitions vary, human-centered approaches to AI aim to develop AI systems that serve the needs, improve the conditions, and align with the values of human stakeholders.…”

“…This question of what do users need to understand about AI systems is core to the nascent field of Human-Centered Explainable AI (HCXAI) [21,22,52], which is a subset of the fields of human centered AI and human centered data science [6,7,25,41,[65][66][67]. Our work is informed by a few key lessons from recent work in HCXAI, mostly conducted in the context of discriminative ML (e.g., for decision-support systems).…”

From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AI

“…With an increasing number of XAI methods, the demand grows for suitable XAI evaluation metrics [1,21,29,86,146]. This need is not only recognized by the AI community; also the Human Computing Interaction (HCI) community is concerned with developing transferable evaluation methods for XAI [65]. In addition, a research agenda for Hybrid Intelligence [6] has explicitly formulated a research question asking how the quality and strength of explanations can be evaluated.…”

Perceive, Understand & Predict - Empirical Indication for Facets in Subjective Information Processing Awareness

“…Quantifying the experience of users cooperating with intelligent systems is a challenging task and many different instruments emerge to support XAI research [41]. Empirically validated tools are crucial in order to 1) research explainable artificial intelligence (XAI) as well as 2) improve transparency for users of, e.g., machine learning models [10]. While tackling the disadvantages of opaque systems, many approaches aim to improve conditions for joint human-AI activity [23] by providing, e.g., visual explanations of the system's information processing [5].…”