Can reinforcement learning agents be e-teachers? Application of machine learning in education
[ 1 ] Instytut Informatyki, Wydział Informatyki i Telekomunikacji, Politechnika Poznańska | [ S ] student | [ P ] employee
2022
chapter in monograph / paper
english
- reinforcement learning
- education
- e-learning
EN E-learning systems are ubiquitous across the whole world. These systems depend on teachers who manually assign tasks to be solved by students in order to optimally develop their skills. However, this is possible only when a teacher runs a course with a small number of students. In large groups teachers are not able to track learning needs of each student. Supporting human teachers with machine learning-based recommendations may allow providing fully personalized learning experience regardless of course size. This paper investigates the possibility of using reinforcement learning algorithms to provide such support. We prepared an environment simulating an e-learning platform. Then we performed an experiment, in which a reinforcement learning agent sets tasks to students to develop their skills. Our experiments show that machine learning algorithms are able to efficiently teach students to increase their level of skill proficiency.
195 - 198
CC BY (attribution alone)
publisher's website
final published version
at the time of publication
20