My research focuses on Artificial Intelligence as an autonomous creator or as a facilitator of human creativity. His work includes computationally intelligent tools for game design, as well as computational creators that blend semantics, visuals, sound, plot and level structure to create horror games, adventure games and more. To accomplish co-creative software, I look at existing methods of computer-aided design and enhancing them by making the computer more proactive, generating suggestions or increasing the level-of-detail of the user's creation. To accomplish that, content is generated procedurally either to maximize a quantifiable measure of game-specific (or level-specific, or artistic) aesthetics or to maximize the novelty of the content pool. Since most game content must fulfill a number of criteria regarding playability, balance and often quality, constrained optimization methods have been developed to allow for a more efficient exploration of a segmented search space containing both feasible and infeasible content. Beyond mixed-initiative tools, my academic interests include fully-automated procedural content generation, digital aesthetics, computational creativity and evolutionary computation. I have also been involved in designing and developing games for a variety of purposes and audiences.
The entries below provide some more insight into the different topics that my research tries to tackle. Each entry contains a short overview of the publications relevant to this line of research. An overview of all publications can be found in list form here.
Procedural Content Generation
Whether generating spaceships, game levels, dungeons or gameworlds, a major part of my research lies within the general area of search-based procedural game content generation through genetic algorithms.
A core goal of my research is to inject some creativity on the part of the computer and become an artificial designer contributing its own ideas and providing feedback and advice similarly to a colleague of the human designer.
Unlike other forms of art, games are bound by playability concerns. Game content is often divided between playable and unplayable. My research focuses on satisfying playability constraints either via objective-driven search or via novelty search.
Divergent search targets unseen or unexpected solutions rather than maximizing an expressly stated objective. For the purposes of my research, I experiment with divergent search for open-ended game content generation as it can break designer fixations and create surprising content.
Several of my research projects aim to discover the preferences and visual taste of a designer from interactions with interactive evolutionary systems, learning the underlying factors of users' choices in order to better accommodate them.
The player's emotional reactions to in-game events, generated content, or suggestions of a computational co-creator are paramount to the players' enjoyment of a game as well as to designers' creativity. I have looked into computational models of affect, such as frustration, tension and others.
Whether to construct predictive models of user behavior or player affect, to find mappings between game elements and their aesthetics, or to figure out the goals of a human co-creator, machine learning based on neural networks is a powerful tool for realizing artificial intelligence.
A vast amount of information about the real-world is available today in knowledge bases and open data repositories. My latest research revolves around extracting such real-world data and finding patterns which can be used to generate complete games.
"Game aesthetics" is a broad and ambiguous term which encompasses believability, challenge, surprise, novelty, visual and aural appeal, interestingness, and more. Examining the "human" aspects of games from a computational perspective is of particular interest to my research.