Michael Mateas

IJCAI Conference 2017 Conference Paper

CHARDA: Causal Hybrid Automata Recovery via Dynamic Analysis

• Adam Summerville
• Joseph Osborn
• Michael Mateas

We propose and evaluate a new technique for learning hybrid automata automatically by observing the runtime behavior of a dynamical system. Working from a sequence of continuous state values and predicates about the environment, CHARDA recovers the distinct dynamic modes, learns a model for each mode from a given set of templates, and postulates \textit{causal} guard conditions which trigger transitions between modes. Our main contribution is the use of information-theoretic measures (1)~as a cost function for data segmentation and model selction to penalize over-fitting and (2)~to determine the likely causes of each transition. CHARDA is easily extended with different classes of model templates, fitting methods, or predicates. In our experiments on a complex videogame character, CHARDA successfully discovers a reasonable over-approximation of the character's true behaviors. Our results also compare favorably against recent work in automatically learning probabilistic timed automata in an aircraft domain: CHARDA exactly learns the modes of these simpler automata.

AAAI Conference 2012 Conference Paper

Learning from Demonstration for Goal-Driven Autonomy

• Ben Weber
• Michael Mateas
• Arnav Jhala

Goal-driven autonomy (GDA) is a conceptual model for creating an autonomous agent that monitors a set of expectations during plan execution, detects when discrepancies occur, builds explanations for the cause of failures, and formulates new goals to pursue when planning failures arise. While this framework enables the development of agents that can operate in complex and dynamic environments, implementing the logic for each of the subtasks in the model requires substantial domain engineering. We present a method using case-based reasoning and intent recognition in order to build GDA agents that learn from demonstrations. Our approach reduces the amount of domain engineering necessary to implement GDA agents and learns expectations, explanations, and goals from expert demonstrations. We have applied this approach to build an agent for the real-time strategy game StarCraft. Our results show that integrating the GDA conceptual model into the agent greatly improves its win rate.

AAAI Conference 2012 Conference Paper

Lessons Learned From a Rational Reconstruction of Minstrel

• Brandon Tearse
• Peter Mawhorter
• Michael Mateas
• Noah Wardrip-Fruin

Scott Turner's 1993 Minstrel system was a high water mark in story generation, harnessing the concept of imaginative recall to generate creative stories. Using case based reasoning and an author level planning system, Minstrel models human creative processes. However, the algorithmic and representational commitments made in Minstrel were never subject to principled and quantitative analysis. By rationally reconstructing Minstrel, we are able to investigate Turner's computational model of creativity and learn new lessons about his architecture. We find that Minstrel's original performance was tied to a well groomed case library, but by modifying several components of the algorithm we can create a more general version which can construct stories using a sparser and less structured case library. Through a rational reconstruction of Minstrel, we both learn new architectural and algorithmic lessons about Minstrel’s computational model of creativity as well as make his architecture available to the contemporary research community for further experimentation.

AAMAS Conference 2008 Conference Paper

Another Look at Search-Based Drama Management

• Michael Mateas

IJCAI Conference 2007 Conference Paper

• Peng Zang
• Manish Mehta
• Michael Mateas
• Ashwin Ram

Typically, autonomous believable agents are implemented using static, hand-authored reactive behaviors or scripts. This hand-authoring allows designers to craft expressive behavior for characters, but can lead to excessive authorial burden, as well as result in characters that are brittle to changing world dynamics. In this paper, we present an approach for the runtime adaptation of reactive behaviors for autonomous believable characters. Extending transformational planning, our system allows autonomous characters to monitor and reason about their behavior execution and to use this reasoning to dynamically rewrite their behaviors. In our evaluation, we transplant two characters in a sample tag game from the original world they were written for into a different one, resulting in behavior that violates the author intended personality. The reasoning layer successfully adapts the character's behaviors so as to bring its long-term behavior back into agreement with its personality.

AAMAS Conference 2007 Conference Paper

A Globally Optimal Algorithm for TTD-MDPs

• Sooraj Bhat
• David L. Roberts
• Mark J. Nelson
• Charles L. Isbell
• Michael Mateas

In this paper, we discuss the use of Targeted Trajectory Distribution Markov Decision Processes (TTD-MDPs)–a variant of MDPs in which the goal is to realize a specified distribution of trajectories through a state space–as a general agent-coordination framework.

AAMAS Conference 2007 Conference Paper

Evaluating a Conversation-Centered Interactive Drama

• Manish Mehta
• Steven Dow
• Michael Mateas

There is a growing interest in developing technologies for creating interactive dramas [13, 22]. Evaluating them, however, remains an open research problem. In this paper, we present a method for evaluating the technical and design approaches employed in a conversation-centered interactive drama. This method correlates players' subjective experience during conversational breakdowns, captured using retrospective protocols, with the corresponding AI processing in the input language understanding and dialog management subsystems. The methodology is employed to analyze conversation breakdowns in the interactive drama Façade. We find that the narrative cues offered by an interactive drama, coupled with believable character performance, can allow players to interpretively bridge system limitations and avoid experiencing a conversation breakdown. Further, we find that, contrary to standard practice for task-oriented conversation systems, using shallowly understood information as part of the system output hampers the player experience in an interactive drama.

AAMAS Conference 2006 Conference Paper

Reinforcement Learning for Declarative Optimization-Based Drama Management

• Mark Nelson
• David Roberts
• Charles Isbell
• Michael Mateas

AAAI Conference 2006 Conference Paper

Targeting Specific Distributions of Trajectories in MDPs

• David L. Roberts
• Charles L. Isbell
• Michael Mateas

We define TTD-MDPs, a novel class of Markov decision processes where the traditional goal of an agent is changed from finding an optimal trajectory through a state space to realizing a specified distribution of trajectories through the space. After motivating this formulation, we show how to convert a traditional MDP into a TTD-MDP. We derive an algorithm for finding non-deterministic policies by constructing a trajectory tree that allows us to compute locally-consistent policies. We specify the necessary conditions for solving the problem exactly and present a heuristic algorithm for constructing policies when an exact answer is impossible or impractical. We present empirical results for our algorithm in two domains: a synthetic grid world and stories in an interactive drama or game.

IROS Conference 2002 Conference Paper

Machines with a different calling [robots for social environments]

• Marc Böhlen
• Michael Mateas

This paper presents an argument for including concepts from the social sciences and the arts in the design of robots for intimate social environments. An example of a robot designed as an interactive dining table situated in a restaurant is described and certain aspects of the particular design approach are explained. It is shown how this can help think about ways to integrate robots into cultural spaces.

AAAI Conference 2000 Conference Paper

Generation of Ideologically-Biased Historical Documentaries

• Michael Mateas
• University of Buffalo; Steffi Domike

Terminal Time is a machine that constructs ideologicallybiased documentary histories in response to audience feedback. The audience answers multiple-choice questions via an applause meter. The answers to these questions influence which historical events are chosen from a knowledge base, how these events will be slanted to embody the bias implied in the audience's answers, and how the events will be connected together to form a historical narrative. Terminal Time's architecture consists of a knowledge base and inference engine for querying the knowledge base, ideological goal trees and rhetorical devices which represent the current bias, a natural language generator to turn the constructed history into narrative prose, and an indexed multimedia database used to sequence video against the narration.