With the fast progress of generative AI in recent years, more games are integrating generated content, raising questions regarding how players perceive and respond to this content. To investigate, we ran a mixed-method survey on the games Super Mario Bros. and Sokoban, comparing procedurally generated levels and levels designed by humans to explore how perceptions of the creator relate to players’ overall experience of gameplay. Players could not reliably identify the level’s creator, yet their experiences were strongly linked to their beliefs about that creator rather than the actual truth. Levels believed to be human-made were rated as more fun and aesthetically pleasing. In contrast, those believed to be AI-generated were rated as more frustrating and challenging. This negative bias appeared spontaneously without knowing the levels’ creator and often was based on unreliable cues of “human-likeness.” Our results underscore the importance of understanding perception biases when integrating generative systems into games.

Procedural content generation via machine learning (PCGML) in games involves using machine learning techniques to create game content such as maps and levels.2D tile-based game levels have consistently served as a standard dataset for PCGML because they are a simplified version of game levels while maintaining the specific constraints typical of games, such as being solvable.In this work, we highlight the unique characteristics of game levels, including their structured discrete data nature, the local and global constraints inherent in the games, and the sensitivity of the game levels to small changes in input.We define the robustness of data as a measure of sensitivity to small changes in input that cause a change in output, and we use this measure to analyze and compare these levels to state-of-the-art machine learning datasets, showcasing the subtle differences in their nature.We also constructed a large dataset from four games inspired by popular classic tile-based games that showcase these characteristics and address the challenge of sparse data in PCGML by providing a significantly larger dataset than those currently available.

Expressive range analysis is a visualization-based technique used to evaluate the performance of generative models, particularly in game level generation.It typically employs two quantifiable metrics to position generated artifacts on a 2D plot, offering insight into how content is distributed within a defined metric space.In this work, we use the expressive range of a generator as the conceptual space of possible creations.Inspired by the quality diversity paradigm, we explore this space to generate levels.To do so, we use a constraintbased generator that systematically traverses and generates levels in this space.To train the constraint-based generator we use different tile patterns to learn from the initial example levels.We analyze how different patterns influence the exploration of the expressive range.Specifically, we compare the exploration process based on time, the number of successful and failed sample generations, and the overall interestingness of the generated levels.Unlike typical quality diversity approaches that rely on random generation and hope to get good coverage of the expressive range, this approach systematically traverses the grid ensuring more coverage.This helps create unique and interesting game levels while also improving our understanding of the generator's strengths and limitations.

This work investigates the role of uncertainty in Slay the Spire using an information-theoretic framework.Focusing on the entropy of game paths (which are based on procedurally-generated maps) we analyze how randomness influences player decision-making and success.By examining a dataset of 20,000 game runs, we quantify the entropy of paths taken by players and relate it with their outcomes and skill levels.The results show that victorious runs are associated with higher normalized entropy, suggesting more risktaking.Additionally, higher-skill players tend to exhibit distinct patterns of risk-taking behavior in later game stages.

When generating game levels, it is desirable for them to be completable.Depending on the designer's goals, it may also be desirable to generate levels without softlocks.A softlock is a situation where the player has not won or lost, but cannot make progress toward the goal, and is thus stuck (unless they reset, or possibly lose the level).In this work we present a constraint-based approach to generating 2D tile-based levels that are completable and do not have areas where the player can get stuck.The approach uses a constraint-based reachability categorization of locations during level generation.This categorization can be used to prevent softlocks by ensuring that all locations reachable going forward from the start of the level are also reachable going backward from the goal, unless they are sinks (areas where the player will inevitably lose the level, e.g.falling off the bottom).Using this approach, it is also possible to intentionally generate levels with softlocks and to repair levels to remove softlocks.

Most research in procedural content generation has, understandably, focused on generating levels that are completable—that is, levels where it is possible for a player to complete them. In this work we explore the generation of uncompletable levels and their applications. Building on an existing constraint-based level generator, we add support for constraints that a level’s goal is not reachable from its start. The generator can thus create levels that are similar to completeable levels in many ways (such as local tile patterns), yet are not possible to complete. We then describe several applications of those constraints and the resulting levels, including: qualitatively characterizing what makes levels uncompletable; creating training data for completability classifiers; checking that a generator can only generate completable levels; and generating levels that require the player to use a special move.

As technical computing software, such as MATLAB and SciPy, has gained popularity, ecosystems of interdependent software solutions and communities have formed around these technologies. The development and maintenance of these technical computing ecosystems requires expertise in both software engineering and the underlying technical domain. The inherently interdisciplinary nature of these ecosystems presents unique challenges and opportunities that shape software development practices.

Determining the completability of levels generated by procedural generators such as machine learning models can be challenging, as it can involve the use of solver agents that often require a significant amount of time to analyze and solve levels. Active learning is not yet widely adopted in game evaluations, although it has been used successfully in natural language processing, image and speech recognition, and computer vision, where the availability of labeled data is limited or expensive. In this paper, we propose the use of active learning for learning level completability classification. Through an active learning approach, we train deep-learning models to classify the completability of generated levels for Super Mario Bros., Kid Icarus, and a Zelda-like game. We compare active learning for querying levels to label with completability against random queries. Our results show using an active learning approach to label levels results in better classifier performance with the same amount of labeled data.