In this paper, we discuss motivations for studying interactive narrative in shared, persistent worlds using the established conventions of quest-based MMORPGs.  We present a framework for categorizing the various techniques used in these games according to the interaction between the world model and the quest model .  Using this framework we generalize recent games to present a more dynamic world model, and investigate extensions to the quest model to support storytelling through adaptive quest narratives.

In this paper, we present the Procedural Game Adaptation (PGA) framework: a designer-controlled way to adapt the Changing the dynamics of a video game (i.e., how the dynamics of a given video game during end-user play. When player's actions affect the game world) is a fundamental tool implemented, this framework produces a deterministic, online of video game design. In Pac-Man, eating a power pill allows adaptation agent (called an experience manager (Riedl the player to temporarily defeat the ghosts that pursue et al. 2011)) that automatically performs two tasks: 1) it and threaten her for the vast majority of the game; in Call gathers information about a game's current player, 2) it of Duty 4, taking the perk called "Deep Impact" allows the uses that information to estimate which of several different player's bullets to pass through certain walls without being changes to the game's dynamics will maximize some playerspecific stopped. The parameters of such changes (e.g., how much value (e.g., fun, sense of influence, etc.). the ghosts slow down while vulnerable) are usually determined by the game's designers long before its release, with

This paper describes a gestural approach to interacting with interactive narrative characters that supports co-creativity. It describes our approach using a Microsoft Kinect to created a short scene with an intelligent avatar and an AI-controlled actor. It describes our preliminary user studies and a recommendation for future evaluation.

We address the problem of determining the structure of a set of plot points for an interactive narrative. To do so, we define a formal two-player game where a computer can play with an author to learn the structural representation of the story. This technique will allow for authors unfamiliar, or uncomfortable, with mathematical structures to create the inputs interactive narrative algorithms require. We include the underlying mathematical theory as a foundation of our approach, and characterize it's effectiveness through a series of simulation experiments. Results indicate there is promise in using formal games to aid in authoring interactive narrative structures.

Interactive narrative is a form of storytelling that adapts to actions performed by users who assume the roles of story characters. To date, interactive narratives are built by hand. In this paper, we introduce Scheherazade, an intelligent system that automatically creates an interactive narrative about any topic from crowdsourced narratives. Our system leverages the experience and creativity of humans by crowdsourcing a corpus of linear narrative examples. It then constructs an executable plot graph, which is a knowledge structure that defines the legal space of an interactive narrative, by learning the plot events, execution precedence, and event separations. We demonstrate the system can successfully construct an interactive narrative based on noisy human input.

Building complete interactive narrative systems is hard. Building systems that are satisfying for naïve users is especially hard since small deficiencies in component technologies can easily destroy the experience for a user. In this paper I argue that we can ameliorate some of these technical limitations through careful choice of genre and style, and discuss a number of properties of farce that make it a particularly attractive choice. Then I will describe work in progress on Punch and Judy AI Playset, a system that allows users to explore possible narratives in the Punch and Judy story world.

A bottleneck in interactive storytelling is the authorial burden of writing narrative units, and connecting them to the interactive narrative structure. To address this problem, we present a hybrid approach that combines AI planning and evolutionary optimization in order to generated new plan operators representing possible story actions, within the framework of a planning-based interactive narrative system. We focus our work on inventing plan operators that are useful for contributing to suspenseful interactive stories, using suspense metrics that have been proposed in the literature. We devise an encoding scheme for converting a plan operator into a genetic-algorithm chromosome and vice versa, respecting constraints that are needed for an operator to be well-formed. We discuss the performance of the system, and several examples from preliminary experiments carried out to evaluate the evolved operators.

Computational models for goal recognition hold great promise for enhancing the capabilities of drama managers and director agents for interactive narratives. The problem of goal recognition, and its more general form, plan recognition, have been the subjects of extensive investigation in the AI community. However, relatively little effort has been undertaken to examine goal recognition in interactive narrative. In this paper, we propose a research agenda to improve the accuracy of goal recognition models for interactive narratives using explicit representations of narrative structure inspired by the natural language processing community. We describe a particular category of narrative representations, narrative schemas, that we anticipate will effectively capture patterns of player behavior in interactive narratives and improve the accuracy of goal recognition models.

My research is focused on using human navigation data ingames and simulation to learn motion models from trajectorydata. These motion models can be used to: 1) track the opponent’smovement during periods of network occlusion; 2)learn combat tactics by demonstration; 3) guide the planningprocess when the goal is to intercept the opponent. A trainingset of example motion trajectories is used to learn twotypes of parameterized models: 1) a second order dynamicalsteering model or 2) the reward vector for a Markov DecisionProcess. Candidate paths from the model serve as themotion model in a set of particle filters for predicting the opponent’slocation at different time horizons. Incorporating theproposed motion models into game bots allows them to customizestheir tactics for specific human players and functionas more capable teammates and adversaries.

The prevalence of virtual worlds presents an interesting The research questions we are looking to solve are: challenge for intelligent mobile agents. Online, very largescale, - How to represent maps of large scale, complex environments persistent virtual worlds such as Second Life (Linden Research Inc. 2012) and massively multi-player online games (MMOs) are becoming more popular. As these - How an agent can generate, update and use these maps worlds grow in size there is a challenge in providing intelligent - How can we utilise user-generated information to build agents that can generate and use maps of these environments and improve upon these maps without the need for hard-coding or pre-processing the map.