Teaching the Foundations of Machine Learning with Candy

Anonymous

Abstract: Machine learning is ubiquitous in decision-making processes across society. The presence and development of ML drives a need for improved education in key concepts at the secondary and tertiary levels that not only trains people to become informed citizens but also trains future researchers to be both principled and ethical practitioners. In this vein, we present a structured classroom activity that simultaneously teaches both supervised classification and critically thinking about ML applications and ethics. We use an active, object-based learning approach to teach supervised classification using a variety of candies, and a problem-based scenario to encourage critical questions about ethics in ML applications

Keywords: Machine Learning, Education, Undergraduate, Secondary Education, Active-Learning

TL;DR: We present an active-learning strategy to teach foundational machine learning skills using candy.