Neural models are one of the most popular approaches for music generation, yet there aren't standard large datasets tailored for learning music directly from game data. To address this research gap, we introduce a novel dataset named NES-VMDB, containing 98,940 gameplay videos from 389 NES games, each paired with its original soundtrack in symbolic format (MIDI). NES-VMDB is built upon the Nintendo Entertainment System Music Database (NES-MDB), encompassing 5,278 music pieces from 397 NES games. Our approach involves collecting long-play videos for 389 games of the original dataset, slicing them into 15-second-long clips, and extracting the audio from each clip. Subsequently, we apply an audio fingerprinting algorithm (similar to Shazam) to automatically identify the corresponding piece in the NES-MDB dataset. Additionally, we introduce a baseline method based on the Controllable Music Transformer to generate NES music conditioned on gameplay clips. We evaluated this approach with objective metrics, and the results showed that the conditional CMT improves musical structural quality when compared to its unconditional counterpart. Moreover, we used a neural classifier to predict the game genre of the generated pieces. Results showed that the CMT generator can learn correlations between gameplay videos and game genres, but further research has to be conducted to achieve human-level performance.

Staple 1990s yoof TV show The Word has just finished with a raucous live performance by some up-and-coming grunge band and now it's time to play video games. In the decade of the original PlayStation and the Sega Saturn, there was no online multiplayer – if you wanted to compete against human beings, you did it in your living room with friends, and anyone else you found in the pub at closing time. It had to be something accessible, something competitive, something that allowed two or even four people to play at once. It needed to have short rounds, because everyone wanted to play. Invariably that would mean one of two options: a footie sim or a fighting game.

I am too much of a control freak to let another player screw up my good work. But I really wanted to try It Takes Two because, first, it was in every single top games of 2021 list and, second, the game is about a couple on the verge of divorce who must find a way to work together. And a little over a year ago, my wife and I were in the same situation. In It Takes Two, the spouses become tiny dolls who must work their way through their suddenly gigantic house, solving puzzles to reunite with their weeping daughter. In real life, we did family therapy.

This paper addresses the problem of evaluating the quality of game levels across different games and even genres, which is of key importance for making procedural content generation and assisted game design tools more generally applicable. Three game design patterns are identified for having high generality while being easily quantifiable: area control, exploration and balance. Formulas for measuring the extent to which a level includes these concepts are proposed, and evaluation functions are derived for levels in two different game genres: multiplayer strategy game maps and single-player roguelike dungeons. To illustrate the impact of these evaluation functions, and the similarity of impact across domains, sets of levels for each function are generated using a constrained genetic algorithm. The proposed measures can easily be extended to other game genres.

Reasoning using expressive symbolic representations is a central theme of AI research, yet there are surprisingly few deployed games, even within the AIIDE research community, that use this sort of "classical" AI. This is partly due to practical and methodological issues, but also due to fundamental mismatches between current game genres and classical AI systems. I will argue that if we want to build games that leverage high-end classical AI techniques like commonsense reasoning and natural language processing, we will also have to develop new game genres and mechanics that better exploit those capabilities. I will also present a design sketch of a game that explores potential game mechanics for classical AI.

Representing games through their pixels offers a promising approach for building general-purpose and versatile game models. While games are not merely images, neural network models trained on game pixels often capture differences of the visual style of the image rather than the content of the game. As a result, such models cannot generalize well even within similar games of the same genre. In this paper we build on recent advances in contrastive learning and showcase its benefits for representation learning in games. Learning to contrast images of games not only classifies games in a more efficient manner; it also yields models that separate games in a more meaningful fashion by ignoring the visual style and focusing, instead, on their content. Our results in a large dataset of sports video games containing 100k images across 175 games and 10 game genres suggest that contrastive learning is better suited for learning generalized game representations compared to conventional supervised learning. The findings of this study bring us closer to universal visual encoders for games that can be reused across previously unseen games without requiring retraining or fine-tuning.

Personalized gamification explores knowledge about the users to tailor gamification designs to improve one-size-fits-all gamification. The tailoring process should simultaneously consider user and contextual characteristics (e.g., activity to be done and geographic location), which leads to several occasions to tailor. Consequently, tools for automating gamification personalization are needed. The problems that emerge are that which of those characteristics are relevant and how to do such tailoring are open questions, and that the required automating tools are lacking. We tackled these problems in two steps. First, we conducted an exploratory study, collecting participants' opinions on the game elements they consider the most useful for different learning activity types (LAT) via survey. Then, we modeled opinions through conditional decision trees to address the aforementioned tailoring process. Second, as a product from the first step, we implemented a recommender system that suggests personalized gamification designs (which game elements to use), addressing the problem of automating gamification personalization. Our findings i) present empirical evidence that LAT, geographic locations, and other user characteristics affect users' preferences, ii) enable defining gamification designs tailored to user and contextual features simultaneously, and iii) provide technological aid for those interested in designing personalized gamification. The main implications are that demographics, game-related characteristics, geographic location, and LAT to be done, as well as the interaction between different kinds of information (user and contextual characteristics), should be considered in defining gamification designs and that personalizing gamification designs can be improved with aid from our recommender system.

This is not a representation of what 2020 felt like -- it's a screen shot from Dead Cells. This is not a representation of what 2020 felt like -- it's a screen shot from Dead Cells. And thank goodness for that, right? Amid worldwide shutdowns, strenuous conversations about police reform, and an endless election cycle, we could all use a break. Do what I do: Pick up your Switch (or whatever console you use) and give yourself a well-deserved, news-free distraction.