Explainable Active Learning (XAL)

Abstract

The wide adoption of Machine Learning (ML) technologies has created a growing demand for people who can train ML models. Some advocated the term "machine teacher'' to refer to the role of people who inject domain knowledge into ML models. This "teaching'' perspective emphasizes supporting the productivity and mental wellbeing of machine teachers through efficient learning algorithms and thoughtful design of human-AI interfaces. One promising learning paradigm is Active Learning (AL), by which the model intelligently selects instances to query a machine teacher for labels, so that the labeling workload could be largely reduced. However, in current AL settings, the human-AI interface remains minimal and opaque. A dearth of empirical studies further hinders us from developing teacher-friendly interfaces for AL algorithms. In this work, we begin considering AI explanations as a core element of the human-AI interface for teaching machines. When a human student learns, it is a common pattern to present one's own reasoning and solicit feedback from the teacher. When a ML model learns and still makes mistakes, the teacher ought to be able to understand the reasoning underlying its mistakes. When the model matures, the teacher should be able to recognize its progress in order to trust and feel confident about their teaching outcome. Toward this vision, we propose a novel paradigm of explainable active learning (XAL), by introducing techniques from the surging field of explainable AI (XAI) into an AL setting. We conducted an empirical study comparing the model learning outcomes, feedback content and experience with XAL, to that of traditional AL and coactive learning (providing the model's prediction without explanation). Our study shows benefits of AI explanation as interfaces for machine teaching--supporting trust calibration and enabling rich forms of teaching feedback, and potential drawbacks--anchoring effect with the model judgment and additional cognitive workload. Our study also reveals important individual factors that mediate a machine teacher's reception to AI explanations, including task knowledge, AI experience and Need for Cognition. By reflecting on the results, we suggest future directions and design implications for XAL, and more broadly, machine teaching through AI explanations.

Are Explanations Helpful? A Comparative Study of the Effects of Explanations in AI-Assisted Decision-Making

“…Reliance and Its Appropriateness We measure the reliance of participants on the AI system via two metrics [22,50]: User Experience We measured participants' perceived autonomy [13], mental demand [21,30,33], perceived complexity [8], engagement [6], future use [30,31], satisfaction [8,19], perceived helpfulness [7,9,28], trust [48], self-efficacy [23] via 7-point Likert scales. The detailed questions are shown in Table 1.…”

How to Evaluate Trust in AI-Assisted Decision Making? A Survey of Empirical Methodologies

“…[33] introduced their own questionnaire, which accounts for cultural influences on trust. 3 papers [67,150,243] combine multiple existing and validated trust questionnaires to create a new one for their studies.…”

Evaluating human-computer interaction education based on active learning methods using the CRITIC and TOPSIS approach

“…The advantages of AI explanation as interfaces for machine instructing-aiding trust calibration and allowing ironic forms of instructing response, along with potential limitations anchoring effect with the model decision and surplus cognitive capability were presented by the study. The study outcomes show significant factors that facilitate a machine instructor's response to AI details, which consists of task information, AI experience, and prerequisite for perception [7]. For virtual higher education systems, the interactive teaching framework utilizing human-machine interaction was analyzed in the study [8].…”

“…By applying formula (6) to the given above decision matrix, we obtain the normalized matrix as given in table 5. The weighted normalized matrix is calculated through the utilization of equation (7) to the above normalized matrix. We apply formula (7) to obtain the weighted normalized matrix and the criteria weights attained through CRITIC technique is as displayed in table 6.…”

“…The weighted normalized matrix is calculated through the utilization of equation (7) to the above normalized matrix. We apply formula (7) to obtain the weighted normalized matrix and the criteria weights attained through CRITIC technique is as displayed in table 6. In this step, we obtain the values of the separation measure and performance score by using equations (8) to (10).…”