We propose the Interactive Constrained MAP-Elites, a quality-diversity solution for game content generation, implemented as a new feature of the Evolutionary Dungeon Designer (EDD): a mixed-initiative co-creativity tool for designing dungeons. The feature uses the MAP-Elites algorithm, an illumination algorithm that segregates the population among several cells depending on their scores with respect to different behavioral dimensions. Users can flexibly and dynamically alternate between these dimensions anytime, thus guiding the evolutionary process in an intuitive way, and then incorporate suggestions produced by the algorithm in their room designs. At the same time, any modifications performed by the human user will feed back into MAP-Elites, closing a circular workflow of constant mutual inspiration. This paper presents the algorithm followed by an in-depth evaluation of the expressive range of all possible dimension combinations in several scenarios, and discusses their influence in the fitness landscape and in the overall performance of the procedural content generation in EDD.

MAP-Elites has been successfully applied to the generation of game content and robot behaviors. However, its behavior and performance when interacted with in co-creative systems is underexplored. This paper analyzes the implications of synthetic interaction for the stability and adaptability of MAP-Elites in such scenarios. We use pre-recorded human-made level design sessions with the Interactive Constrained MAP-Elites (IC MAP-Elites). To analyze the effect of each edition step in the search space over time using different feature dimensions, we introduce Temporal Expressive Range Analysis (TERA). With TERAs, MAP-Elites is assessed in terms of its adaptability and stability to generate diverse and high-performing individuals. Our results show that interactivity, in the form of design edits and MAP-Elites adapting towards them, directs the search process to previously unexplored areas of the fitness landscape and points towards how this could improve and enrich the co-creative process with quality-diverse individuals.

We propose the use of quality-diversity algorithms for mixed-initiative game content generation. This idea is implemented as a new feature of the Evolutionary Dungeon Designer, a system for mixed-initiative design of the type of levels you typically find in computer role playing games. The feature uses the MAP-Elites algorithm, an illumination algorithm which divides the population into a number of cells depending on their values along several behavioral dimensions. Users can flexibly and dynamically choose relevant dimensions of variation, and incorporate suggestions produced by the algorithm in their map designs. At the same time, any modifications performed by the human feed back into MAP-Elites, and are used to generate further suggestions.

The Evolutionary Dungeon Designer (EDD) [1] is as a mixed-initiative tool for creating dungeons for adventure games. Results from a user study with game developers positively evaluated EDD as a suitable framework for collaboration between human designers and PCG suggestions, highlighting these as time-saving and inspiring for creating dungeons [2]. Previous work on EDD identified the need of assessing aesthetic criteria as a key area for improvement in its PCG Engine. By upgrading the individual encoding system and the fitness evaluation in EDD's evolutionary algorithm, we present three techniques to preserve and account the designer's aesthetic criteria during the dungeon generation process: the capability of locking sections for preserving custom aesthetic structures, as well as the measurement of symmetry and similarity in the provided suggestions.

Mixed-initiative systems highlight the collaboration between humans and computers in fostering the generation of more interesting content in game design. In light of the ever-increasing cost of game development, providing mixed-initiative tools can not only significantly reduce the cost but also encourage more creativity amongst game designers. The Evolutionary Dungeon Designer (EDD) [3] is a mixed-initiative tool with a focus on using evolutionary computation to procedurally generate content that adhere to game design patterns. As part of an ongoing project, feedback from a user study on EDD's capabilities as a mixed-initiative design tool pointed out the need for improvement on the tool's functionalities [4]. In this paper we present a review of the principles of the mixed-initiative model, as well as the existing approaches that implement it. The outcome of this analysis allows us to address the appointed needs for improvement by shaping a new version of EDD that we describe here. Finally, we also present the results from a user study carried out with professional game developers, in order to assess EDD's new functionalities. Results show an overall positive evaluation of the tool's intuitiveness and capabilities for empowering game developers' creative skills during the design process of dungeons for adventure games. They also allow us to identify upcoming challenges pattern-based mixed-initiative tools could benefit from.

Abstract: Procedural Content Generation (PCG) can be a useful tool for aiding creativity in the process of designing game levels. Mixed-initiative level generation tools where a designer and an algorithm collaborate to iteratively generate game levels have been used for this purpose. However, it can be difficult for designers to work with tools that do not respond to the common language of games: game design patterns. We present the Evolutionary Dungeon Designer, the first step towards a mixed-initiative dungeon design tool which evolves dungeon rooms using game design patterns, as well as several metrics regarding the placement of treasures and enemies, in the fitness function of a genetic algorithm. Our results show that we are able to control the frequency, shape and type of design patterns, as well as properly place enemies and treasures in the generated rooms, using design pattern-related input parameters.

Mixed-initiative Procedural Content Generation uses algorithms to assist human designers in the collaborative creation of game content. Different mixed-initiative approaches use different methods to engage with the design material while supporting the designer’s intentions. However, the designer runs the risk of misunderstanding the system’s abilities and how to control them. In order to limit miscommunication during the design process, heuristics could be applied. In this paper we present a mixed-initiative tool for evolving dungeons with the aid of game design patterns as heuristics. The tool, the Evolutionary Dungeon Designer, uses a genetic algorithm that searches for levels containing game design patterns on two hierarchical levels of abstraction to express more complex gameplay in the game level. We evaluate the tool through a series of lab experiments and a user study conducted with professional game developers. Our results demonstrate that we are able to control the generation of the different patterns with the aid of design pattern-related input parameters, as well as identifying a number of features a design pattern-based mixed-initiative tool could benefit from.

The development of content in digital games, such as game worlds, quests, levels, 3D-models, and textures, is costly and time consuming. To address this, different approaches to automate the process of creating game content, often referred to as procedural content generation (PCG), has been suggested. However, PCG is a complex task and include challenges such as creating content with variation, coherent style, speed, and correctness. The research in the thesis is concerned with generating game content with the aid of game design patterns, both by establishing models and exploring different methods to generate actual game content for different games. The methods include implementations of evolutionary computation, i.e. a set of search-based approaches that searches for instances of game design patterns on different abstraction levels that make up Super Mario Bros. (SMB) levels and a learning algorithm implementation based on a model (n-grams) of patterns from the original SMB-game. The different generators were evaluated with metrics concerned with the expressive range of the generators and with user tests.

This paper analyses dungeons, of the varieties commonly found in role-playing games, into several sets of design patterns at different levels of abstraction. The analysis focuses on mechanical patterns that could be either straightforwardly instantiated or recognized by a well-defined process. At the most concrete level a set of fundamental components were identified, followed by a long list of micropatterns which can be directly instantiated. Shorter lists of meso- and macro-patterns, which can be identified mechanically, are also identified. The direct motivation for this analysis is to find building blocks and objectives for a search-based procedural dungeon generator, however we believe the analysis can be useful for understanding this class of game artifacts in general. In particular, the constraints on patterns being instantiable or recognizable leads to a stricter pattern analysis than many other attempts at analyzing game design.

Evaluation is an open problem in procedural content generation research. The eld is now in a state where there is a glut of content generators, each serving di erent purposes and using a variety of techniques. It is difficult to understand, quantitatively or qualitatively, what makes one generator di erent from another in terms of its output. To remedy this, we have conducted a large-scale comparative evaluation of level generators for the Mario AI Benchmark, a research-friendly clone of the classic platform game Super Mario Bros. In all, we compare the output of seven different level generators from the literature, based on different algorithmic methods, plus the levels from the original Super Mario Bros game. To compare them, we have de ned six expressivity metrics, of which two are novel contributions in this paper. These metrics are shown to provide interestingly di erent characterizations of the level generators. The results presented in this paper, and the accompanying source code, is meant to become a benchmark against which to test new level generators and expressivity metrics.

Generating content at multiple levels of abstraction simultaneously is an open challenge in procedural content generation. Representing and automatically replicating the style of a human designer is another. This paper addresses both of these challenges through extending a previously devised methodology for pattern-based level generation. This method builds on an analysis of Super Mario Bros levels into three abstraction levels: micro-, meso- and macro-patterns. Micro-patterns are then used as building blocks in a search-based PCG approach that searches for macro-patterns, which are defined as combinations of meso-patterns. Results show that we can successfully generate levels that replicate the macro-patterns of selected input levels, and we argue that this constitutes an approach to automatically analysing and replicating style in level design.

We shown that novel, linear game levels can be created using n- grams that have been trained on a corpus of existing levels. The method is fast and simple, and produces levels that are recognisably in the same style as those in the corpus that it has been trained on. We use Super Mario Bros. as an example domain, and use a selection of the levels from the original game as a training corpus. We treat Mario levels as a left-to-right sequence of vertical level slices, allowing us to perform level generation in a setting with some formal similarities to n-gram-based text generation and music generation. In empirical results, we investigate the effects of corpus size and n (sequence length). While the applicability of the method might seem limited to the relatively narrow domain of 2D games, we argue that many games in effect have linear levels and n-grams could be used to good effect, given that a suitable alphabet can be found.

In this paper we present a search-based approach for procedural generation of game levels that represents levels as sequences of micro-patterns and searched for meso-patterns. The micro-patterns are “slices” of original human-designed levels from an existing game, whereas the meso-patters are abstractions of common design patterns seen in the same levels. This method generates levels that are similar in style to the levels from which the original patterns were extracted, while still allowing for considerable variation in the geometry of the generated levels. The evolutionary method for generating the levels was tested extensively to investigate the distribution of micro-patterns used and meso-patterns found.

In this paper we present a search-based approach for procedural generation of game levels that represents levels as sequences of micro-patterns and searched for meso-patterns. The micro-patterns are "slices" of original human-designed levels from an existing game, whereas the meso-patters are abstractions of common design patterns seen in the same levels. This method generates levels that are similar in style to the levels from which the original patterns were extracted, while still allowing for considerable variation in the geometry of the generated levels. The evolutionary method for generating the levels was tested extensively to investigate the distribution of micro-patterns used and meso-patterns found.

Rooted in the complexity of purposeful design, this paper embraces a phenomenological perspective of design as both a process and artifact. We use this perspective to interpret why the conceptualization and realization of design intentions can be difficult to achieve and why design is often perceived as a so called ‘wicked problem’. This paper revisits the concepts of game design patterns and game mechanics, arguing that refactoring these concepts is needed to clarify their relationships and motivations. We outline the separation of concerns between them and suggest that an additional contextualizing layer should be added to the discourse. Using this, we define and reflect upon what we refer to as the conceptual relationship model.

This paper discusses how to use design patterns in procedural level generation, with particular reference to the classic console game Super Mario Bros. In a previous paper, we analyzed the levels in this game to nd a set of recurring level design patterns, and discussed an implementation where levels were produced from concatenation of these patterns. In this paper, we instead propose using patterns as design objectives. An implementation of this based on evolutionary computation is presented. In this implementation, levels are represented as a set of vertical slices from the original game, and the tness function count the number of patterns found. Qualitative analysis of generated levels is performed in order to identify strengths and challenges of this method.

Procedural content generation and design patterns could potentially be combined in several di erent ways in game design. This paper discusses how to combine the two, using automatic platform game level design as an example. The paper also present work towards a pattern-based level generator for Super Mario Bros, namely an analysis of the levels of the original Super Mario Bros game into 23 di erent patterns.

Massively multiplayer online games have become a common research subject in game studies. Several of these studies have focused on how the player interacts with the game and other players through the game, but often the fact is neglected that there are other games besides MMOGs that allow players to interact with each other. Two-player off-line games, for example, also allow for interaction between players, both through the game and in the physical world. This chapter focuses on the improvement of skills while such a game is being played. We use interaction analysis to understand how the player learns to play the game and how to play together with someone else. By observing how players interact with each other and with the game in a setting with only two players, we find different learning situations than one would find in a single-player or MMOG environment. These learning situations show that the formulation of an understanding of a game and the incorporation of the game into the player’s repertoire are obstructed by the repertoire itself, and they show that players may have trouble adapting to a reflective playing style. This case study is part of a larger project on situated play with multiplayer off-line and colocated video games.

Are typical computer game genres still valid descriptors and useful for describing game structure and game content? Games have changed from simple to complex and from single function to multi function. By identifying structural differences in game elements we develop a more nuanced model to categorized games and use cluster analysis as a descriptive tool in order to do so. The cluster analysis of 75 functionally different games shows that the two perspectives (omnipresent and vagrant), as well as challenges, mutability and savability are important functional categories to use in order to describe games.