Comparative Analysis of GAT and BERT for Human-Like Playtesting

2026-07-13

AI Summary

This work addresses the limitations of existing data-driven gameplay testing models, which struggle to comprehensively capture player strategies and require frequent redesign of feature engineering and network architectures when new game mechanics are introduced. To overcome these challenges, the authors propose a general-purpose representation framework that, for the first time, integrates BERT with Graph Attention Networks (GAT) to predict human-like gameplay in puzzle games such as Candy Crush Saga. By explicitly modeling the relational structure of the game board, the approach significantly reduces reliance on handcrafted features. Experimental results demonstrate that the proposed method substantially outperforms CNN-based baselines on complex board configurations, yielding improved generalization and more accurate simulation of player behavior.

Abstract

Accurately modeling and understanding player experience is crucial for designing engaging puzzle games. To achieve this, a common approach involves collecting diverse user data to train predictive playtesting models that mimic player behavior. However, existing data-driven methods often lack the ability to capture the full range of player strategies and require extensive feature engineering and network architecture modeling. This limitation becomes particularly evident when new game mechanics or features are introduced, which necessitate continual adjustments to the models. To addrss these challenges, we propose a more generalized representation that reduces - or even eliminates - the need for ongoing feature-engineering maintenance. Specifically, we investigate two general-purpose network architectures: (a) a transformer-based model (BERT) and (b) a graph attention model (GAT), both of which are designed to effectively capture the relational structure of Candy Crush Saga (CCS) game boards. Our experiments compare these approaches to Convolutional Neural Networks (CNN) baselines, revealing better performance on challenging board configurations and underscoring the benefits of our generalizable representation.

Problem

• player experience modeling
• feature engineering
• generalizable representation
• playtesting
• game mechanics

Innovation

• BERT
• GAT
• generalizable representation
• playtesting
• graph attention network