Search-based procedural content generation methods allow video games to introduce new content continually, thereby engaging the player for a longer time while reducing the burden on developers. However, games so far have not explored the potential economic value of unique evolved artifacts. Building on this insight, this paper presents for the first time a Facebook game called Petalz in which players can share flowers they breed themselves with other players through a global marketplace. In particular, the market in this social game allows players to set the price of their evolved aesthetically-pleasing flowers in virtual currency. Furthermore, the transaction in which one player buys seeds from another creates a new social element that links the players in the transaction. The combination of unique user-generated content and social gaming in Petalz facilitates meaningful collaboration between users, positively influences the dynamics of the game, and opens new possibilities in digital entertainment.

The creative nature of games makes trying new ideas desirable, but these changes are sometimes very risky. We need to find ways to minimize risks while we build innovative experiences. Driver San Francisco did this by using Hinted-execution Behavior Trees; this technique allows developers to modify existing AI behaviors dynamically with very low risk, and was used to adapt Driver’s getaway AI to players’ skills.

We present POMCoP, a system for online planning in collaborative domains that reasons about how its actions will affect its understanding of human intentions, and demonstrate its use in building sidekicks for cooperative games. POMCoP plans in belief space. It explicitly represents its uncertainty about the intentions of its human ally, and plans actions which reveal those intentions or hedge against its uncertainty. This allows POMCoP to reason about the usefulness of incorporating information gathering actions into its plans, such as asking questions, or simply waiting to let humans reveal their intentions. We demonstrate POMCoP by constructing a sidekick for a cooperative pursuit game, and evaluate its effectiveness relative to MDP-based techniques that plan in state space, rather than belief space.

Automated sports commentary is a form of automated narrative. Sports commentary exists to keep the viewer informed and entertained. One way to entertain the viewer is by telling brief stories relevant to the game in progress. We introduce a system called the Sports Commentary Recommendation System (SCoReS) that can automatically suggest stories for commentators to tell during games. Through several user studies, we compared commentary using SCoReS to three other types of commentary and show that SCoReS adds significantly to the broadcast across several enjoyment metrics. We also collected interview data from professional sports commentators who positively evaluated a demonstration of the system. We conclude that SCoReS can be a useful broadcast tool, effective at selecting stories that add to the enjoyment and watchability of sports. SCoReS is a step toward automating sports commentary and, thus, automating narrative.

Game balancing is the fine-tuning phase in which a functioning game is adjusted to be deep, fair, and interesting. Balancing is difficult and time-consuming, as designers must repeatedly tweak parameters, and run lengthy playtests to evaluate the effects of these changes. If designers could receive immediate feedback on their designs, they could explore a vast space of variations, and select only the most promising games for playtesting. Such automated design feedback has been difficult to achieve, as there is no mathematical formulation of game balance that unifies many of its forms. We argue for a formulation in which carefully restricted agents are played against standard agents. We develop this restricted-play balance framework, and evaluate its utility by building a tool capable of calculating measures of balance for a large family of games. By applying this tool to an educational card game, we demonstrate how the framework and tool allow designers to rapidly evaluate and iterate on the balance of their games.

A common approach to interactive narrative involves imbuing the computer with all of the potential story pre-authored story experiences (e.g. as beats, plot points, planning operators, etc.). This has resulted in an accepted paradigm where stories are not created by or with the user; rather, the user is given piecemeal access to the story from the gatekeeper of story knowledge: the computer (e.g. as an AI drama manager). This article describes a formal process that provides for the equal co-creation of story-rich experiences, where neither the user nor computer is in a privileged position in an interactive narrative. It describes a new formal approach that acts as a first step for the real-time co-creation of narrative in games that rely on the negotiated shared mental model between a human actor and an AI improv agent.

Automating parts of game creation benefits both professional and amateur game designers and much previous work has already made progress on this front. In this paper we tackle automating level design. We describe a general graph-based representation for game levels and present a preliminary system that leverages this representation. Our system automatically explores existing levels of a 2D platform game using the rapidly-exploring random tree (RRT) algorithm and constructs a compact graph representation from this exploration. Our system can also modify a graph representation on-the-fly to reflect user-directed changes to the existing level structure. This work constitutes an initial step toward the larger goal of automating level design in a general way.

TRANSIT is a fast and optimal technique for computing shortest path costs in road networks. It is attractive for its usually modest memory requirements and impressive running times. In this paper we give a first analysis of TRANSIT routing on a set of popular grid-based video-game benchmarks taken from the AI pathfinding literature. We show that in the presence of path symmetries, which are inherent to most grids but normally not road networks, TRANSIT is strongly and negatively impacted, both in terms of performance and memory requirements. We address this problem by developing a new general symmetry breaking technique which adds small random epsilon-values to edges in the search graph, reducing the size of the TRANSIT network by up to 4 times while preserving optimality. Using our enhancements TRANSIT achieves up to four orders of magnitude speed improvement vs. A* search and uses in many cases only a small (<=10MB) or modest (<= 50MB) amount of memory. We also compare TRANSIT with CPDs, a recent and very fast database-driven pathfinding approach. We find the algorithms have complementary strengths but also identify a class of problems for which TRANSIT is up to two orders of magnitude faster than CPDs using a comparable amount of memory.

The goal of predictive sparse coding is to learn a representation of examples as sparse linear combinations of elements from a dictionary, such that a learned hypothesis linear in the new representation performs well on a predictive task. Predictive sparse coding algorithms recently have demonstrated impressive performance on a variety of supervised tasks, but their generalization properties have not been studied. We establish the first generalization error bounds for predictive sparse coding, covering two settings: 1) the overcomplete setting, where the number of features k exceeds the original dimensionality d; and 2) the high or infinite-dimensional setting, where only dimension-free bounds are useful. Both learning bounds intimately depend on stability properties of the learned sparse encoder, as measured on the training sample. Consequently, we first present a fundamental stability result for the LASSO, a result characterizing the stability of the sparse codes with respect to perturbations to the dictionary. In the overcomplete setting, we present an estimation error bound that decays as \tilde{O}(sqrt(d k/m)) with respect to d and k. In the high or infinite-dimensional setting, we show a dimension-free bound that is \tilde{O}(sqrt(k^2 s / m)) with respect to k and s, where s is an upper bound on the number of non-zeros in the sparse code for any training data point.

Feature selection is a technique to screen out less important features. Many existing supervised feature selection algorithms use redundancy and relevancy as the main criteria to select features. However, feature interaction, potentially a key characteristic in real-world problems, has not received much attention. As an attempt to take feature interaction into account, we propose L1-LSMI, an L1-regularization based algorithm that maximizes a squared-loss variant of mutual information between selected features and outputs. Numerical results show that L1-LSMI performs well in handling redundancy, detecting non-linear dependency, and considering feature interaction.