Nested Wave Function Collapse Enables Large-Scale Content Generation

Procedural Content Generation (PCG) has emerged as a powerful approach for automating game content creation, offering significant benefits in terms of cost reduction and time efficiency compared to traditional game design and development processes (Zhang, Zhang, & Huang, 2022). While Genetic Algorithms (GAs) have been widely used in PCG, alternative metaheuristic algorithms such as Particle Swarm Optimization (PSO) and Artificial Bee Colony (ABC) have demonstrated their effectiveness in delivering high-quality solutions and efficient optimization capabilities across different problem domains (Amato, 2017). However, their application in PCG remains limited. I aim to evaluate the performance of PSO and ABC in map layout generation, challenging the conventional use of GAs. By comparing three metaheuristic algorithms (GA, ABC, and PSO) I seek to assess the effectiveness of these approaches in generating game levels and identify any obvious differences in their performance characteristics. Comprehensive experiments are conducted, applying GA, ABC, and PSO to a map layout generation. Metrics like convergence speed and content quality are used to evaluate the generated game content. My findings reveal that both ABC and PSO demonstrate advantages over traditional GA implementations when generating game levels, indicating their potential for enhancing PCG. In this presentation, I will share the results of comparing three metaheuristic algorithms (GA, PSO, and ABC) in map layout generation for game levels, emphasizing the potential benefits of leveraging diverse algorithmic approaches to create more captivating and immersive game worlds. Also, I will conclude with a call for further research in this area to expose new possibilities in content generation. By considering varied metaheuristic approaches, game developers can improve content generation techniques and create more captivating and interactive player experiences.

Procedural content generation (PCG) systems are designed to automatically generate content for video games. PCG for physics-based puzzles requires one to simulate the game to ensure feasibility and stability of the objects composing the puzzle. The major drawback of this simulation-based approach is the overall running time of the PCG process, as the simulations can be computationally expensive. This paper introduces a method that uses machine learning to reduce the number of simulations performed by an evolutionary approach while generating levels of Angry Birds, a physics-based puzzle game. Our method uses classifiers to verify the stability and feasibility of the levels considered during search. The fitness function is computed only for levels that are classified as stable and feasible. An approximation of the fitness that does not require simulations is used for levels that are deemed as unstable or unfeasible by the classifiers. Our experiments show that naively approximating the fitness values can lead to poor solutions. We then introduce an approach in which the fitness values are approximated with the average fitness value of the levels' parents added to a penalty value. This approximation scheme allows the search procedure to find good-quality solutions much more quickly than a competing approach—we reduce from 43 to 25 minutes the running time required to generate one level of Angry Birds.

Tools for Landscape Analysis of Optimisation Problems in Procedural Content Generation for Games

Procedural content generation (PCG)-based game design aims to reach a new way of playing games by focusing gameplay around algorithmic game content generation. However, positioning interaction with PCG systems and generated content to the center of player experience poses design challenges for both game design and AI design. In order to create the wanted affordances, rich contextual information is required to make informed decisions on the generated content. While previous research has presented excellent developments on PCG’s possibilities, further considering context and affordances in the early stages of prototyping may aid designers reach these possibilities in a more consistent manner. This study is set to discuss how context, affordances and the game’s overall design can be considered during the prototyping process of PCG-based games. Misaligned game context and affordances can result in deeply rooted design issues that may later manifest as subpar gameplay experiences and increased development effort. These emergent issues are examined through a post-mortem case study to produce an extended PCG-based design process, featuring actionable steps, that takes context, affordances, and the game’s overall design into account through meaningful play.

Procedural content generation has become a popular research topic in recent years. However, most content generation systems are specialized to a single game. We are interested in methods that can generate content for a wide variety of games without a game-specific algorithm design. Statistical approaches are a promising avenue for such generators and, more specifically, map generators. In this paper, we explore Markov models as a means of modeling and generating content for multiple domains. We apply our Markov models to Super Mario Bros. , Loderunner , and Kid Icarus in order to determine how well our models perform in terms of the playability of the content generated, the expressive ranges of the models, and the effects of training data on those expressive ranges.

This study takes a new perspective on the procedural content generation (PCG) evaluation problem, extracts current PCG evaluation methods from previous works, and presents a novel classification of these methods while showing each method’s capabilities. Also, the present study introduces a novel concept called Panda Evaluation. Additionally, the soft and hard launches were presented as two evaluation methods and possible building blocks of PE. A group of papers was analyzed to understand previous works and find new opportunities. In doing so, some missing PCG evaluation areas were found, and some new methods were proposed for future PCG evaluations. To the best of our knowledge, this is the first time these concepts have been presented in PCG evaluation.

Case-based learning (CBL) is an educational method widely used in health professional training, emphasizing theoretical knowledge’s practical application. However, traditional CBL has many challenges related to the complexity and accessibility of case scenarios and its demands on educators to effectively facilitate and evaluate student discussions. Despite its popularity and proven benefits, the comparative effectiveness and efficiency of CBL against newer educational technologies still need to be explored. In this quasi-experimental study, third-year physiotherapy students at the University of Lleida used a computer program for procedural content generation (PCG) and interactive clinical cases (ICs), and assessed them using the Spanish physiotherapy self-efficacy scale and the System Usability Scale, and a satisfaction scale. The study showed a significant improvement in self-efficacy among participants after using PCG and ICs. The usability of the PCG was moderate, while the ICs scored higher and had greater user satisfaction. Satisfaction metrics revealed a strong preference among students for incorporating clinical cases into other subjects, with higher satisfaction levels observed in the IC group compared to the PCG group. The study demonstrated that PCG and ICs significantly enhance clinical skills acquisition in physiotherapy education by improving student performance and engagement.

This paper describes the design of a novel approach to procedural content generation, intent on supporting game design activities. The distinctive factor in this approach is that content generation is guided by a series of target experience indicators, which the designer can define freely according to his own agenda. We detail its underlying concepts and procedural logic, as well as its purported benefits, and outline early experiments in the design process of a prototype.

Although game‐based learning has been increasingly promoted in education, there is a need to adapt game content to individual needs for personalized learning. Procedural content generation (PCG) offers a solution for difficulty in developing game contents automatically by algorithmic means as it can generate individually customizable game contents applicable to various objectives. In this paper, we advanced a data‐driven PCG approach benefiting from a genetic algorithm and support vector machines to automatically generate educational‐game contents tailored to individuals' abilities. In contrast to other content generation approaches, the proposed method is not dependent on designer's intuition in applying game contents to fit a player's abilities. We assessed this data‐driven PCG approach at length and showed its effectiveness by conducting an empirical study of children who played an educational language‐learning game to cultivate early English‐reading skills. To affirm the efficacy of our proposed method, we evaluated the data‐driven approach against a heuristic‐based approach. Our results clearly demonstrated two things. First, users realized greater performance gains from playing contents tailored to their abilities compared with playing uncustomized game contents. Second, this data‐driven approach was more effective in generating contents closely matching a specific player‐performance target than the heuristic‐based approach.

Procedural Content Generation is applied in the development process of many commercial games: automatically generated game contents are delivered to players in order to offer a constantly changing user experience and enrich the game itself. Usually, the generative process relies on search-based non-deterministic algorithms, which encode one or more techniques for guaranteeing “legal” yet diversified output. Declarative approaches to content generation, more properly defined as Declarative Content Specification techniques, like the ones based on Answer Set Programming, allow to focus on describing content requirements rather than programming ad-hoc generation engines, and to fast prototype generation techniques themselves. This work investigates to what extent ASP-based DCS is scalable enough for industrial contexts, by proposing a partitioning-based approach. A working prototype, available as an Unity Asset and as a GVGAI framework level generator is presented.

Procedural content generation (PCG) is a method of content creation fully or semi-performed by computers. PCG is widely used in game development to generate game content, from Rogue (1998) to No Man's Sky (2016). PCG generates final contents, which are ready to be added to a game, or intermediate contents, which are might work as content sketch to be polished by human designers. In this paper we survey the current state of procedural dungeon generation (PDG) research, a subarea of PCG. We analyzed the works according to the game features they generate, the solution strategy employed and the taxonomy of procedural content generation. Some of the relevant findings of the survey are: (1) PDG for 3D levels has been little explored; (2) few works supported levels with barriers, a game mechanic which blocks, temporarily, the player progression, and; (3) and just a few solutions relied on mixed-initiative approach, where a human design content is combined with a computer generated level.

Procedural Content Generation (PCG) techniques enable automatic creation of diverse and complex environments. While PCG facilitates more efficient content creation, ensuring consistently high-quality, industry-standard content remains a significant challenge. In this research, we propose a method to identify and repair unstable levels generated by existing PCG models. We use Angry Birds as a case study, demonstrating our method on game levels produced by established PCG approaches. Our method leverages object segmentation and visual analysis of level images to detect structural gaps and perform targeted repairs. We evaluate multiple object segmentation models and select the most effective one as the basis for our repair pipeline. Experimental results show that our method improves the stability and playability of AI-generated levels. Although our evaluation is specific to Angry Birds, our image-based approach is designed to be applicable to a wide range of 2D games with similar level structures.

Educational games are powerful tools in the era of Education 4.0. However, their applications are hampered by many issues, including high development costs. Procedural content generation (PCG) and medium coupling are two potent methods for reducing an educational game's development cost; however, their joint application is under-studied. In this study, we explored the topic through a PCG application for generating placements of objects representing elements of learning content in a game map. The content consisted of correct answers, which the player must gather in the correct order, and wrong answers, which the player must avoid. We employed an evolutionary algorithm in three stages of the generation: it first generated a minimal set of correct answer element objects (CAEOs), followed by generating a copy of the set, and then a set of wrong answer element objects (WAEOs). We employed and tested a fitness function that calculated the mean and standard deviation of object pair distances. The results show that the generation algorithm is capable of generating CAEOs and WAEOs distributions that are unpredictable and encourage exploration. We find that multiplying the fitness function's standard deviation variable by a specific value, which depends on how many objects are to be generated, is crucial to the algorithm's success. We also discuss the limitations of this study and directions for future ones.

Procedural content generation is helping game developers to create significant quantity of high quality dynamic content for video games at a fraction of cost of the traditional methods. Procedural texture synthesis is a sub category of procedural content generation which helps video games to have significant variations in textures of the environments and the objects across the progress of the game and to avoid repetition. Generative Adversarial Networks are a class of deep learning algorithms which are capable of learning the patterns and creating new patterns. In this paper, Generative Adversarial Networks is used for procedural content generation for original textures synthesis for video game development. This method is used by video game designers for autonomous redesigning of objects and environment textures. This process saves significant time and cost in video game development. The particular attention in this paper is on procedural synthesis of ground surface textures. The generated texture samples are visually acceptable and have mean score of 2.45 with 0.1 standard deviation after 2K iteration. Also the discriminator loss of generated samples reached 0.74 at the final stage of training. The proposed framework can be used as an effective procedural texture synthesis framework for video game design and development.

For high skilled players, an easy game might become boring and for low skilled players, a difficult game might become frustrating. This research's goal is to offer players a personalized experience adapted according to their performance and levels of attention. We created a simple side-scrolling 2D platform game using Procedural Content Generation, Dynamic Difficulty Adjustment techniques and brain computer data obtained from players in real time using an Electroencephalography device. We conducted a series of experiments with different players and got results that confirm that our method is adjusting each level according to performance and attention.