We present CLASSQL, a multi-agent model for playing real-time strategy games, where learning and control of our own team's units is decentralized; each agent uses its own reinforcement learning process to learn and control units of the same class. Coordination between these agents occurs as a result of a common reward function shared by all agents and synergistic relations in a carefully crafted state and action model for each class.

In this paper we describe a method of modeling play styles as deviations from approximations of game theoretically rational actions. These deviations are interpreted as containing information about player skill and player decision making style. We hypothesize that this information is useful for differentiating between players and for understanding why human player behavior is attributed intentionality which we argue is a prerequisite for believability. To investigate these hypotheses we describe an experiment comparing 400 games in the Mario AI Benchmark testbed, played by humans, with equivalent games played by an approximately game theoretically rationally playing AI agent. The player actions' deviations from the rational agent's actions are subjected to feature extraction, and the resulting features are used to cluster play sessions into expressions of different play styles. We discuss how these styles differ, and how believable agent behavior might be approached by using these styles as an outset for a planning agent. Finally, we discuss the implications of making assumptions about rational game play and the problematic aspects of inferring player intentions from behavior.

This article describes a technical approach to movement-based interactions between a human interactor and an intelligent agent based on the theatrical Viewpoints movement framework. The installation creates a liminal virtual / real space for the human and AI to interact by the use of digital projection for the AI visualization and shadow play to represent the human. Observations from a recent public demonstration of the system and future directions of work are also discussed.

Collaboration between agents and players within games is a ripe area for exploration. As with adversarial AI, collaborative agents are challenged to accurately model players and adapt their behavior accordingly. The task of cooperation, however, allows for communication between teammates that can prove beneficial in coordinating joint actions and plans. Furthermore, we propose extending established multi-agent communication paradigms to include transfer of information pertinent to player models. By querying goal and preference information from a player, an agent can reduce uncertainty in coordination domains, allowing for more effective planning. We discuss the challenges as well as the planned development and evaluation of the system.

Coordination with an unknown human teammate is a notable challenge for cooperative agents. Behavior of human players in games with cooperating AI agents is often sub-optimal and inconsistent leading to choreographed and limited cooperative scenarios in games. This paper considers the difficulty of cooperating with a teammate whose goal and corresponding behavior change periodically. Previous work uses Bayesian models for updating beliefs about cooperating agents based on observations. We describe belief models for on-line planning, discuss tuning in the presence of noisy observations, and demonstrate empirically its effectiveness in coordinating with inconsistent agents in a simple domain. Further work in this area promises to lead to techniques for more interesting cooperative AI in games.

This paper examines the performance of a number of AI agents on the games included in the General Video Game Playing Competition. Through analyzing these results, the paper seeks to provide insight into the strengths and weaknesses of the current generation of video game playing algorithms. The paper also provides an analysis of the given games in terms of inherent features which define the different games. Finally, the game features are matched with AI agents, based on performance, in order to demonstrate a plausible case for algorithm portfolios as a general video game playing technique.

Real-Time Strategy (RTS) games have become an increasingly popular test-bed for modern artificial intelligence techniques. With this rise in popularity has come the creation of several annual competitions, in which AI agents (bots) play the full game of StarCraft: Broodwar by Blizzard Entertainment. The three major annual StarCraft AI Competitions are the Student StarCraft AI Tournament (SSCAIT), the Computational Intelligence in Games (CIG) competition, and the Artificial Intelligence and Interactive Digital Entertainment (AIIDE) competition. In this paper we will give an overview of the current state of these competitions, and the bots that compete in them.

Video games presents an immense dissonance between its narrative and interactive sections. These two elements are commonly presented in a format where each occurs at a certain time, but rarely simultaneously. Besides, other elements, such as the background music, are also usually pre-defined, denying for the users the condition to establish new forms of creative expression within the system. This work, therefore, intends to propose new systems and algorithms to fit this paradigm.

My thesis aims at conceptualizing and implementing a computational model of narrative generation that is informed by narratological theory as well as cognitive multi-agent simulation models. It approaches this problem by taking a mimetic stance towards fictional characters and investigates how narrative phenomena related to characters can be computationally recreated from a deep character model grounded in multi agent systems. Based on such a conceptualization of narrative it explores how the generation of plot can be controlled, and how the quality of the resulting plot can be evaluated, in dependence of fictional characters. By that it contributes to research on computational creativity by implementing an evaluative storytelling system, and to narratology by proposing a generative narrative theory based on several post-structuralist descriptive theories.

Over the past few years the Angry Birds AI competition has been held in an attempt to develop intelligent agents that can successfully and efficiently solve levels for the video game Angry Birds. Many different agents and strategies have been developed to solve the complex and challenging physical reasoning problems associated with such a game. However, the performance of these various agents is non-transitive and varies significantly across different levels. No single agent dominates all situations presented, indicating that different procedures are better at solving certain levels than others. We therefore propose the construction of a hyper-agent that selects from a portfolio of sub-agents whichever it believes is best at solving any given level. This hyper-agent utilises key features that can be observed about a level to rank the available candidate algorithms based on their expected score.The proposed method exhibits a significant increase in performance over the individual sub-agents, and demonstrates the potential of using such an approach to solve other physics-based games or problems.