SangGyu Nam

Dr. SangGyu Nam

orcid

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 September 2021 in IEEE Transactions on Games
Verified
Journal Article
Closed access
SNSangGyu Nam
SangGyu Nam
Chu-Hsuan HsuehChu-Hsuan Hsueh
Chu-Hsuan Hsueh
KIKokolo Ikeda
Kokolo Ikeda
Many recent studies in procedural content generation (PCG) are based on machine learning. One of the promising approaches is generative models, which have shown impressive results in generating new pictures and videos from existing ones. However, it is usually costly to collect sufficient content for training on PCG. To address this issue, we consider reinforcement learning (RL), which does not need to collect training data in advance but learns from its interaction with an environment. In this work, RL agents are trained to generate stages, which we define as series of events in turn-based role playing games. It is a challenging task since several events in a stage are usually highly correlated to each other. We first formulate the stage generation problem into a Markov decision process. A hand-crafted evaluation function, which simulates players’ enjoyment, is defined to evaluate generated stages. Two RL algorithms are selected in the experiments, which are deep Q-network for discrete action space and deep deterministic policy gradient for continuous action space. The generated stages from both models receive evaluation values indicating good quality. To solve the delayed reward problem and further improve the quality of the stages, we employ virtual simulations (VS) to give rewards to intermediate actions and get stages with higher average scores. In addition, we introduce noise to avoid generating similar stages while trying to keep the quality as high as possible. The proposed methods succeed in generating good and diverse stages.

1 August 2019
Verified
Conference Paper
Closed access
SNSangGyu Nam
SangGyu Nam
KIKokolo Ikeda
Kokolo Ikeda
In this study, procedural content generation (PCG) using reinforcement learning (RL) is focused. PCG is defined as the generation of game content tailored to the defined evaluation function using RL models, which is one of the examples of PCG via machine learning. Compared to other generation content areas such as computer vision and natural language process, generative models such as variational autoencoders, PixelCNN, and generative adversarial networks exhibit some difficulties for applications to the game area because during the development of a new game, the content data used for training is typically not sufficient. Hence, RL is considered to be used as a method for PCG. In particular, the stage of turn-based RPG is selected as our research target because it comprises discrete sections, and its parameters were closely related; hence, it is a challenge to generate desirable stages, and the main goal is to generate various stages guided by the designed evaluation function. Two RL models, Deep Q-Network and Deep Deterministic Policy Gradient, respectively, are selected, and the generated stages are evaluated as 0.78 and 0.85 by our designed function, respectively. By the application of the stochastic noise policy, diverse stages are successfully obtained, and those diversities are evaluated by the parameter mse and the different number of valid strategies.

1 January 2025 in Journal of Information Technology Education: Research
Verified
Journal Article
Open access
PDF file
AGAnderson Kevin Gwenhure
Anderson Kevin Gwenhure
SNSangGyu Nam
SangGyu Nam
Aim/Purpose: To evaluate the sustainability of gamified cybersecurity education, training, and awareness (ETA) initiatives by addressing recurring limitations, conceptual misclassifications, and the overlooked influence of novelty and duration. Background: Gamification has seen widespread application in cybersecurity ETA initiatives and is frequently credited with improving engagement, motivation, and learning outcomes. However, its true effectiveness remains uncertain. Methodology: A systematic literature review (SLR) following the PRISMA framework was conducted, analyzing 12 peer-reviewed empirical studies focused on gamified cybersecurity ETA interventions. Findings: The review reveals that most gamified initiatives are short-term and assessed within the novelty effect window, which may inflate their perceived effectiveness. Only two studies applied established gamification frameworks, highlighting a widespread reliance on improvised designs. Common elements like leaderboards and time pressure often cause unintended negative effects, such as anxiety and disengagement. Additionally, poor reporting on intervention duration and negative outcomes hinders reproducibility. Long-term behavior change remains largely unexamined. By synthesizing these findings, this study offers design guidance and calls for more structured, evidence-based approaches to gamification in cybersecurity. Recommendations for Practitioners: Design for long-term impact, not short-term stimulation. Align game elements with user motivations and context. Use theory-based frameworks to document positive and negative outcomes to support continuous improvement and behaviour change. Recommendation for Researchers: Researchers should adopt longitudinal study designs to assess the sustainability of gamified interventions beyond the novelty phase, focusing on behavioural change rather than short-term engagement. Future studies must explore optimal durations for meaningful outcomes and examine how game elements affect diverse user groups and contexts. Clarifying distinctions between gamification, serious games, and simulations is essential to reduce conceptual ambiguity. Developing context-sensitive frameworks that incorporate motivational and environmental factors is also critical. Finally, standardizing reporting on duration, design models, and both positive and negative outcomes will improve comparability and advance research in this field. Future Research: This research advances the design of evidence-based gamified interventions to help close the cybersecurity skills gap and promote a more security-conscious digital culture. Organizational adoption can enhance user awareness and reduce human-related risks. Future research should distinguish gamification from related methods like serious games and simulations, and shift focus from short-term engagement to long-term outcomes such as secure behavior and compliance. Studies must report both positive and negative effects, considering demographic and contextual factors. Developing integrated, context-aware frameworks and conducting longitudinal studies to determine optimal intervention durations are essential. Standardized reporting and exploration of gamified internal policies will further support effective and sustainable cybersecurity awareness.

28 June 2023
Verified
Conference Paper
Closed access
SNSangGyu Nam
SangGyu Nam
CHChu-Hsuan Hsueh
Chu-Hsuan Hsueh
KIKokolo Ikeda
Kokolo Ikeda
Procedural content generation (PCG) is one of the major research topics in the game research field. The objective of PCG is to automatically generate reusable game content, such as levels, items, or landscapes, that satisfy specific criteria such asp layability, believability, or difficulty. To the best of our knowledge, this paper i s the first to propose a level generation approach for Super Mario Bros. (Super Mario) that considers natural connection. In this study, a Super Mario level consists of several sublevels called patterns. Without natural connectivity of patterns, the level may be unplayable or not entertaining. We employ Conditional Generative Adversarial Networks (CGAN) to generate naturally connected patterns with a Matching mechanism to repair the defective outputs obtained from CGAN. The experimental results indicate that the levels generated with the proposed CGAN and Matching technique significantly improve natural connectivity compared to randomly generated patterns. The findings c an help improve the PCG's capability to generate content for better players' enjoyment.

19 June 2024 in IEEE Transactions on Games
Verified
Journal Article
Closed access
SNSangGyu Nam
SangGyu Nam
Chu-Hsuan HsuehChu-Hsuan Hsueh
Chu-Hsuan Hsueh
PRPavinee Rerkjirattikal
Pavinee Rerkjirattikal
KIKokolo Ikeda
Kokolo Ikeda
Procedural content generation (PCG) is essential in game development, automating content creation to meet various criteria such as playability, diversity, and quality. This paper leverages reinforcement learning (RL) for PCG to generate Super Mario Bros levels. We formulate the problem into a Markov decision process (MDP), with rewards defined using player enjoyment-based evaluation functions. Challenges in level representation and difficulty assessment are addressed by conditional generative adversarial networks (CGAN) and human-like AI agents that mimic aspects of human input inaccuracies. This ensures that the generated levels are appropriately challenging from human perspectives. Furthermore, we enhance content quality through virtual simulation (VS), which assigns rewards to intermediate actions to address a credit assignment problem (CAP). We also ensure diversity through a diversity-aware greedy policy (DAGP), which chooses not-bad-but-distant actions based on Q-values. These processes ensure the production of diverse and high-quality Super Mario levels. Human subject evaluations revealed that levels generated from our approach exhibit natural connection, appropriate difficulty, non-monotony, and diversity, highlighting the effectiveness of our proposed methods. The novelty of our work lies in the innovative solutions we propose to address challenges encountered in employing PCGRL in Super Mario Bros, contributing to the field of PCG for game development.