A Reinforcement Learning Framework for Personalized Adaptive E-Learning

Personalized learning is motivated by the recognition that students show diverse learning styles and paces due to factors such as personality characteristics, motivation, emotional and environmental circumstances, and prior experiences. It is also increasingly important to account for students with conditions such as Attention-Deficit/Hyperactivity Disorder (ADHD) or other learning intervening factors. Accommodating individual differences in attention span and learning patterns is crucial for effective learning. When designing a digital course, many parameters can be adapted to the unique learner profile such as presentation style of the content, stimuli for enhanced attention, length of session, available links, assessment and navigation options and more. This chapter suggests the use of Reinforcement Learning (RL) algorithm for a personalized digital learning experience, linking the learner’s profile with the responses of the learning environment. We suggest a framework, based on Universal Design Learning (UDL) principles, where an intelligent agent is programmed to learn the student’s learning skills and preferences, then adapt to the user by offering suitable learning materials, structures and stimuli, accounting for continuous changes in performance. A simulation is presented to validate the adaptive algorithm applied to a digital course, focused but not limited to the parameters relevant to students with ADHD such as attention and distractibility.