Development of an Interactive User Interface for Teaching and Analysing the Pebble Game
No Thumbnail Available
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Nottingham
Abstract
The Pebble Game, widely utilised within mathematical and computational research, acts as a vital
tool for operationalising complex principles derived from finite model theory. This game provides
a robust framework for exploring the interplay between formal languages and finite structures,
addressing core issues in computational complexity. Increasing interest in finite model theory
reflects its profound impact on the advancement of theoretical knowledge and its application in
resolving practical challenges.
This study explores the design and implementation of a user interface for the Pebble Game, which
aims to demystify and enhance the understanding of complex concepts in graph theory and
computational logic. The project integrates a Q-learning algorithm to simulate strategic gameplay,
providing an automated and adaptable opponent. We designed the user interface (UI) to improve
the game's accessibility and engagement, particularly for learners from diverse educational
backgrounds. Our research focused on analysing the total rewards generated by the AI across
multiple gaming sessions, a critical metric for assessing the adopted techniques' efficacy. The
results demonstrated the AI's capacity for knowledge acquisition and gradual improvement in
gameplay, underscoring the utility of reinforcement learning in educational tools.
The results indicate that the improved Pebble Game interface not only helps achieve the
educational goals of simplifying complicated theoretical concepts but also offers a strong
foundation for future enhancements, such as integrating user feedback and refining the AI's
learning algorithms. This study contributes significantly to the broader field of educational
technology by illustrating how the synergy between interactive design and artificial intelligence
can boost learning outcomes.
Based on the study's results, we suggest implementing a user feedback mechanism to
continuously refine the user interface, along with customising the game's difficulty level based on
each user's progress. Lastly, we recommend conducting long-term studies to evaluate the game's
impact on academic achievement over time.
Description
Keywords
Finite model theory, the pebble game, reinforcement learning, Q-learning.