Learning effective policies in multi-agent adversarial games is a significant challenge since the search space can be prohibitively large when the actions of all the agents are considered simultaneously. Recent advances in Monte Carlo search methods have produced good results in single-agent games like Go with very large search spaces. In this paper, we propose a variation on the Monte Carlo method, UCT (Upper Confidence Bound Trees), for multi-agent, continuous-valued, adversarial games and demonstrate its utility at generating American football plays for Rush Football 2008. In football, like in many other multi-agent games, the actions of all of the agents are not equally crucial to gameplay success. By automatically identifying key players from historical game play, we can focus the UCT search on player groupings that have the largest impact on yardage gains in a particular formation.

Poor or nonexistent team coordination of non-player characters in games can break the immersive experience for players. When an AI agent is surrounded by allies but behaves as though it is alone, the result is an unbelievable situation. Team coordination allows characters to work together and produce better tactics and strategy, which generates better behavior. Characters acting alone are limited in their space of action; the lone sniper in an action game can find a hidden position offering an excellent view of an ambush area, but without teammates to drive the enemy to the ambush, the sniper may end up being useless. A worse problem is when Figure 1: Teams of agents in the FI3RST environment the game provides AI characters as allies for the player but the allies hinder the player.

In this paper, we show that personalized levels can be auto- matically generated for platform games. We build on previ- ous work, where models were derived that predicted player experience based on features of level design and on playing styles. These models are constructed using preference learn- ing, based on questionnaires administered to players after playing different levels. The contributions of the current pa- per are (1) more accurate models based on a much larger data set; (2) a mechanism for adapting level design parameters to given players and playing style; (3) evaluation of this adap- tation mechanism using both algorithmic and human players. The results indicate that the adaptation mechanism effectively optimizes level design parameters for particular players.

Artificial intelligence is frequently used to control virtual characters in movies and games. When these characters appear in crowds, controlling them is called crowd simulation. In this paper, I suggest that crowd simulation could be accomplished by multi-agent reinforcement learning, a method by which groups of agents can learn to act autonomously in their environment. I present a case study that explores the challenges and benefits of this type of approach and encourages the development of learning techniques for AI in entertainment media.

One of the main challenges in game AI is building agents that can intelligently react to unforeseen game situations. In real-time strategy games, players create new strategies and tactics that were not anticipated during development. In order to build agents capable of adapting to these types of events, we advocate the development of agents that reason about their goals in response to unanticipated game events. This results in a decoupling between the goal selection and goal execution logic in an agent. We present a reactive planning implementation of the Goal-Driven Autonomy conceptual model and demonstrate its application in StarCraft. Our system achieves a win rate of 73% against the built-in AI and outranks 48% of human players on a competitive ladder server.

We have developed and demonstrated an experimental authoring and run-time tool, called Disco for Games, that supports the creation of games in which dialogue and action are integrated without the need for changing modes. This tool is based on collaborative discourse theory and hierarchical task networks, in which utterances are treated as actions, and has a number of additional benefits including better modeling of interruptions, automatic dialogue generation, plan recognition and automatic failure retry.

Real-time pedestrian simulation for open-world games involves aggressive behavior simplification and culling to keep computational cost under control, but it is diffficult to predict whether these techniques will become unrealistic in certain situations. We propose a method of perceptually simulating highly realistic pedestrian behavior in virtual cities in realtime. Designers build a highly realistic simulation, from which a perceptually identical "perceptual simulation" is generated. Although the perceptual simulation simulates only a small portion of the world at a time, and does so with inexpensive approximations, it can be statistically guaranteed that the results are perceptually indistinguishable from those of the original simulation.

This paper proposes an automatic model-based approach that enables adaptive decision making in modern virtual games. It builds upon the Integrated MDP and POMDP Learning AgeNT (IMPLANT) architecture which has shown to provide plausible adaptive decision making in modern games. However, it suffers from highly time-consuming manual model specification problems. By incorporating an automated priority sweeping based model builder for the MDP, as well as using the Tactical Agent Personality for the POMDP, the work in this paper aims to resolve these problems. Empirical proof of concept is shown based on an implementation in a modern game scenario, whereby the enhanced IMPLANT agent is shown to exhibit superior adaptation performance over the old IMPLANT agent whilst eliminating manual model specifications and at the same time still maintaining plausible speeds.

Game environments are becoming more and more mutable from the actions of both Players and Non Player Characters (NPCs). However, current generation AI agents do not take advantage of the tactical abilities these mutable worlds provide. We propose a method to make the game agents aware of the mutability of the world by extending their repertoire of abilities to include world alteration commands and some evaluation functions, which determine when and where to alter the world for the greatest tactical gain. Primarily, our work focuses on the Adversarial Navigation Mesh Alteration (ANMA) algorithm, which evaluates potential changes to the map in adversarial environments from an attacker and defender point of view. We present an empirical evaluation of the ANMA algorithm in a Capture The Flag (CTF) simulation environment with several teams of agents. One group of agents (adaptive) lacks the ability to initiate world deformations, but they can respond and re-plan to take advantage of world modifications. The second team of agents (builders) can only generate additional paths through the world using the attacker portion of ANMA. The third team of agents (universal) is able to fully deform the world by generating new paths or removing existing paths using both the attacker and defender sections of ANMA. We evaluated these teams and observed that builder agents beat adaptive agents at a rate of 1.33 to 1. The more advanced universal agents beat adaptive agents at a rate of 2.75 to 1 and builder agents 1.4 to 1.

This paper describes our current approach in implementing computational improvisational micro-agents. This approach is intended to foster bottom-up research to better understand how to build more complex agent behaviors in a theatrical improvisational setting. Micro-agent designs are based on our current findings in a multi-year study focused on studying real life theatrical improvisers with an aim towards better understanding the cognition employed inimprovisation at the individual and group level. It also introduces a key architectural component from the domain of fuzzy logic that enables us to clearly represent some of our current findings.