Optimal Hierarchical Learning Path Design with Reinforcement Learning

Abstract

E-learning systems are capable of providing more adaptive and efficient learning experiences for students than the traditional classroom setting. A key component of such systems is the learning strategy, the algorithm that designs the learning paths for students based on information such as the students' current progresses, their skills, learning materials, and etc. In this paper, we address the problem of finding the optimal learning strategy for an E-learning system. To this end, we first develop a model for students' hierarchical skills in the E-learning system. Based on the hierarchical skill model and the classical cognitive diagnosis model, we further develop a framework to model various proficiency levels of hierarchical skills. The optimal learning strategy on top of the hierarchical structure is found by applying a model-free reinforcement learning method, which does not require information on students' learning transition process. The effectiveness of the proposed framework is demonstrated via numerical experiments.

Curiosity-Driven Recommendation Strategy for Adaptive Learning via Deep Reinforcement Learning

“…, s(t − 1)}. Similar learning procedures can be found in Chen et al (2018); Li, Xu, Zhang, and Chang (2018); Tan et al (2019); Tang, Chen, Li, Liu, and Ying (2019).…”

“…) that pushes the learner to master as many knowledge points as possible, where w k is the weight for the k-th knowledge point. Li et al (2018) and Tan et al (2019) added the current learning time t in the reward setting to pursue the learning efficiency. With the learning experience data that contains the sequence of observed states, actions and rewards, extracting a reward intrinsic to the agent seems to be more reasonable to reflect the inner curiosity of the learner.…”

Extending exploratory diagnostic classification models: Inferring the effect of covariates

“…Diagnostic models provide a statistical framework for designing formative assessments by classifying student knowledge profiles according to a collection of fine-grained attributes (de la Torre & Douglas, 2004;Rupp et al, 2010;von Davier, 2008). In fact, learning technology algorithms leverage the diagnostic modelling framework by tailoring learning interventions to adapt to individual students' capabilities and needs (Chen, Li, et al, 2018;Han et al, 2020;Huang et al, 2019;Li et al, 2018;Tan et al, 2020;Zhang & Chang, 2016) and to track skill development Madison & Bradshaw, 2018;Studer, 2012;Wang et al, 2017;Wang, Zhang, et al, 2018;Ye et al, 2016;Yigit & Douglas, 2021;Zhang & Chang, 2020).…”