A contextual bandits framework for personalized learning action selection

Recent developments in machine learning have the potential to revolutionize education by providing an optimized, personalized learning experience for each student. We study the problem of selecting the best personalized learning action that each student should take next given their learning history; possible actions could include reading a textbook section, watching a lecture video, interacting with a simulation or lab, solving a practice question, and so on. We first estimate each student’s knowledge profile from their binary-valued graded responses to questions in their previous assessments using the SPARFA framework. We then employ these knowledge profiles as contexts in the contextual (multi-armed) bandits framework to learn a policy that selects the personalized learning actions that maximize each student’s immediate success, i.e., their performance on their next assessment. We develop two algorithms for personalized learning action selection. While one is mainly of theoretical interest, we experimentally validate the other using a real-world educational dataset. Our experimental results demonstrate that our approach achieves superior or comparable performance as compared to existing algorithms in terms of maximizing the students’ immediate success.