Intelligent interactive narrative systems often use an experience manager to govern the behavior of non-player characters in a way that guides the story towards its author's agenda, which may be for entertainment, education, training, or other purposes. For such systems, a central challenge is creating believable virtual characters. The Belief Desire Intention framework (Bratman 1987) is often cited as a goal for researchers in this field; for characters to seem realistic, a human audience should attribute beliefs, desires, and intentions to them. Much of my prior work has focused on belief; my goal for the future is to finish the work on belief, and to implement a new model of desire and intention that explicitly reasons about characters' commitment to certain plans of action. Experience management has been framed as a plot graph traversal problem which is jointly solved by the player and an AI experience manager (Bates 1992; Weyhrauch 1997).
Rational agents are becoming prevalent in many domains, from data analysis to entertainment and games. The increased prevalence of agents has evolved new tools and techniques to work with and design new agents. One such technique is system simulation. Systems simulation is a technique an author can use to imitate tasks, processes, or systems, and in particular, agents. Systems simulation has a variety of uses, ranging from simulating ecological systems to entertainment, such as interactive narratives and digital games. However, many system simulators use specialized programming languages and require prior programming experience. This causes a disconnect between individuals with limited programming experience who wish to use the simulation tools, and the software itself. New users may find the specialized languages daunting, and the initial learning process too intense for the anticipated reward. This research strives to bridge the gap between system simulation tools and users with little to no programming experience. Future work includes a corpus of narrative and autonomous agent creation tools designed for users with little to no programming experience.
An immersive, interactive environment and the non-player characters populating it often play a key role in interactive narrative experiences. We posit that if a procedurally generated narrative is better able to reflect real-world attributes of the player's surroundings, then the experience would be more transportive for the player. With this comes the problem of generating believable narratives and characters for an open, complex, real world. Simulating such a society within the constraints of the real environment, and allowing for virtual characters to more accurately mimic human behavior could increase the believability of the agents. The interactions amongst these agents sharing their cultural views, biases, and histories based on their real-world geolocation could inform the study of audience modeling and machine enculturation, allowing computers to learn or reason about social norms in regions. Finally, we posit this research would afford better applications in the field of entertainment or computational social science.
Many modern creative industrial processes rely on the collaboration between multiple humans, assisted by one or more computational systems, in a complex environment. However, most traditional systems lack the adaptability required to contribute in a flexible, co-creative manner, instead executing a fixed set of tasks in a preset time schedule. We believe games, especially cooperative games offer an ideal platform to conduct research in co-creativity. We present our motivation, preliminary work and future goals to study, build and measure game-inspired co-creative AI systems.
Samuel, Ben (University of New Orleans) | Reed, Aaron (Spirit AI) | Short, Emily (Spirit AI) | Heck, Samantha (University of Idaho) | Robison, Barrie (University of Idaho) | Wright, Landon (University of Idaho) | Soule, Terence (University of Idaho) | Treanor, Mike (American University) | McCoy, Joshua (University of California, Davis) | Sullivan, Anne (Georgia Institute of Technology) | Shirvani, Alireza (University of New Orleans) | Garcia, Edward (University of New Orleans) | Farrell, Rachelyn (University of New Orleans) | Ware, Stephen (University of New Orleans) | Compton, Katherine (University of California, Santa Cruz)
We describe a tool based on the Wave Function Collapse algorithm that performs example-based path generation on fixed maps. Our design aims at a practical system usable by non-programmers, and includes both easy input control and multiple post-processing steps. The design is implemented in Unity and enables users to easily visualize the results of experimenting with different path descriptions and game levels.
We present a new way to represent and understand experience managers - AI agents that tune the parameters of a running game to pursue a designer's goal. Existing representations of AI managers are diverse, which complicates the task of drawing useful comparisons between them. Contrary to previous representations, ours uses a point of unity as its basis: that every game/manager pair can be viewed as only a game with the manager embedded inside. From this basis, we show that several common, differently-represented concepts of experience management can be re-expressed in a unified way. We demonstrate our new representation concretely by comparing two different representations, Search-Based Drama Management and Generalized Experience Management, and we present the insights that we have gained from this effort.
Within the area of procedural content generation (PCG) there are a wide range of techniques that have been used to generate content. Many of these techniques use traditional artificial intelligence approaches, such as genetic algorithms, planning, and answer-set programming. One area that has not been widely explored is straightforward combinatorial search -- exhaustive enumeration of the entire design space or a significant subset thereof. This paper synthesizes literature from mathematics and other subfields of Artificial Intelligence to provide reference for the algorithms needed when approaching exhaustive procedural content generation. It builds on this with algorithms for exhaustive search and complete examples how they can be applied in practice.
Players of digital games face numerous choices as to what kind of games to play and what kind of game content or in-game activities to opt for. Among these, game content plays an important role in keeping players engaged so as to increase revenues for the gaming industry. However, while nowadays a lot of game content is generated using procedural content generation, automatically determining the kind of content that suits players' skills still poses challenges to game developers. Addressing this challenge, we present matrix- and tensor factorization based game content recommender systems for recommending quests in a single player role-playing game. We discuss the theory behind latent factor models for recommender systems and derive an algorithm for tensor factorizations to decompose collections of bipartite matrices. Extensive online bucket type tests reveal that our novel recommender system retained more players and recommended more engaging quests than handcrafted content-based and previous collaborative filtering approaches.
The ability of digital storytelling agents to evaluate their output is important for ensuring high-quality human-agent interactions. However, evaluating stories remains an open problem. Past evaluative techniques are either model-specific--- which measure features of the model but do not evaluate the generated stories ---or require direct human feedback, which is resource-intensive. We introduce a number of story features that correlate with human judgments of stories and present algorithms that can measure these features. We find this approach results in a proxy for human-subject studies for researchers evaluating story generation systems.