Abstract:

Artificial Intelligence (AI) is experiencing rapid growth in science and publicity. Because of its usefulness, developers and researchers are working to integrate it into everyday life. Most useful AI tools are based on Machine Learning (ML), which represents a subcategory of AI. ML is then divided into supervised, unsupervised, and reinforcement learning, which are differentiated based on the data given and the response provided to the learning agent.

AI has proved itself helpful by simplifying many repetitive tasks for humans. Because of its enormous applicability in various fields, experts are trying to bring it into education, especially the personalization of learning. This paper will present the idea of teaching multiplication tables with the assistance of intelligent tutors, based on Reinforcement Learning, i.e., agents adapt their actions during the learning process according to the responses they receive from the environment. Learning multiplication tables is a continuous interaction between the intelligent tutor (the learning agent) and the learner (the environment). In each interaction, the intelligent tutor assigns a task to the student. The student solves the task and returns the solution to the tutor, who then updates its functioning based on the received solution. The number of interactions during the learning process can be predetermined, or the learning can be time-limited.

The aim is to adapt the intelligent tutor to the learner to maximize the learner's performance. The learner would interact with the intelligent tutor through a simple user interface that would store data about their interactions. This data would then be used to measure the performance of the system and personalize learning, as well as to identify potential problems with the agent. In this paper, we will present preliminary research on a small group of learners.

Keywords:

AI in education, adaptive learning, reinforcement learning, multiplication tables.