Browsing by Subject "Computer games"

Now showing 1 - 5 of 5

Automated domain analysis and transfer learning in general game playing
(2010-08) Kuhlmann, Gregory John; Stone, Peter, 1971-; Lifschitz, Vladimir; Mooney, Raymond J.; Porter, Bruce W.; Schaeffer, Jonathan
Creating programs that can play games such as chess, checkers, and backgammon, at a high level has long been a challenge and benchmark for AI. Computer game playing is arguably one of AI's biggest success stories. Several game playing systems developed in the past, such as Deep Blue, Chinook and TD-Gammon have demonstrated competitive play against the top human players. However, such systems are limited in that they play only one particular game and they typically must be supplied with game-specific knowledge. While their performance is impressive, it is difficult to determine if their success is due to generally applicable techniques or due to the human game analysis. A general game player is an agent capable of taking as input a description of a game's rules and proceeding to play without any subsequent human input. In doing so, the agent, rather than the human designer, is responsible for the domain analysis. Developing such a system requires the integration of several AI components, including theorem proving, feature discovery, heuristic search, and machine learning. In the general game playing scenario, the player agent is supplied with a game's rules in a formal language, prior to match play. This thesis contributes a collection of general methods for analyzing these game descriptions to improve performance. Prior work on automated domain analysis has focused on generating heuristic evaluation functions for use in search. The thesis builds upon this work by introducing a novel feature generation method. Also, I introduce a method for generating and comparing simple evaluation functions based on these features. I describe how more sophisticated evaluation functions can be generated through learning. Finally, this thesis demonstrates the utility of domain analysis in facilitating knowledge transfer between games for improved learning speed. The contributions are fully implemented with empirical results in the general game playing system.
The effects of a problem based learning digital game on continuing motivation to learn science
(2008-05) Toprac, Paul K., 1960-; Liu, Min, Ed. D.
The purpose of this study was to determine whether playing a problem-based learning (PBL) computer game, Alien Rescue III, would promote continuing motivation (CM) to learn science, and to explore the possible sources of CM. Another goal was to determine whether CM and interest to learn science in the classroom were identical constructs. CM was defined as the pursuit of academic learning goals in noninstructional contexts that were initially encountered in the classroom. Alien Rescue was played for a total of 9 hours in the seventh grade of a private middle school with 44 students, total, participating. The study used a design-based research approach that attempted to triangulate quantitative and qualitative methods. A science knowledge test, and two selfreport questionnaires--one measuring motivation and one measuring CM--were administered preintervention, postintervention, and follow-up. Qualitative data was also collected, including student interviews, classroom observations, written responses, and a science teacher interview. Repeated measures ANOVAs were used to determine any significant changes in scores. A multiple regression analysis was used to explore whether a model of CM could be determined using the Eccles’ expectancy-value achievement motivation model. The constant comparative method was used to obtain relevant information from the qualitative data. Based on contradictory quantitative and qualitative findings, results were mixed as to whether students exhibited an increase in CM to learn space science. Students continued to freely engage Alien Rescue during the mid-class break, but this does not strictly adhere to the definition of CM. However, many students did find space science more interesting than anticipated and developed increased desire to learn more in class, if not outside of class. Results also suggest that CM and interest in learning more in class are separate but related constructs. Finally, no satisfactory model emerged from the multiple regression analysis but based on students’ interviews, continuing interest to learn is influenced by all the components of Eccles’ expectancy-value model. Response effects may have confounded quantitative results. Discussion includes challenges of researching in classrooms, CM, and Eccles’ motivational model, and the tension between PBL and game based approaches. Future design recommendations and research directions are provided.
Evolving visibly intelligent behavior for embedded game agents
(2006) Bryant, Bobby Don; Miikkulainen, Risto
Machine learning has proven useful for producing solutions to various problems, including the creation of controllers for autonomous intelligent agents. However, the control requirements for an intelligent agent sometimes go beyond the simple ability to complete a task, or even to complete it efficiently: An agent must sometimes complete a task in style. For example, if an autonomous intelligent agent is embedded in a game where it is visible to human observers, and plays a role that evokes human intuitions about how that role should be fulfilled, then the agent must fulfill that role in a manner that does not dispel the illusion of intelligence for the observers. Such visibly intelligent behavior is a subset of general intelligent behavior: a subset that we must be able to provide if our methods are to be adopted by the developers of games and simulators. This dissertation continues the tradition of using neuroevolution to train artificial neural networks as controllers for agents embedded in strategy games or simulators, expanding that work to address selected issues of visibly intelligent behavior. A test environment is created and used to demonstrate that modified methods can create desirable behavioral traits such as flexibility, consistency, and adherence to a doctrine, and suppress undesirable traits such as seemingly erratic behavior and excessive predictability. These methods are designed to expand a program of work leading toward adoption of neuroevolution by the commercial gaming industry, increasing player satisfaction with their products, and perhaps helping to set AI forward as The Next Big Thing in that industry. As the capabilities of research-grade machine learning converge with the needs of the commercial gaming industry, work of this sort can be expected to expand into a broad and productive area of research into the nature of intelligence and the behavior of autonomous agents.
A task-based approach to teaching Spanish to young language learners using computer games
(2011-05) Vazquez, Sinthia Sarai; Horwitz, Elaine Kolker, 1950-; Martinez-Roldan, Carmen
The abundance of technologies around our children, provides us with resources that can be used in second and foreign language classrooms. Often, children do not have the opportunity to practice Spanish in an authentic way, due to limited Spanish instruction that some institutions or public schools offer at the elementary level. Therefore, the limited time that is allowed to teach should be used wisely by means of computer games in the target language in conjunction with language tasks may offer the opportunity to learn and practice the second language (L2). The purpose of the present report is to: present existing literature on tasks and computer games in foreign/second language learning; suggest how they can be incorporated in a task-based approach in terms of teaching Spanish as an L2 to young learners; show examples of computer games in company with various language tasks that can be used for L2 learning; and provide an example of a lesson plan based on the suggested approach. Also, some of the benefits of this Spanish task-based approach will be discussed. Finally, important teaching implications are offered based on the existing literature on tasks and the task-based approach using computer games that is proposed in the this report.
Using a game template as a multimedia-based cognitive tool to facilitate novices' conceptual understanding of object-oriented programming
(2008-12) Yuen, Timothy Tung-Ming, 1978-; Liu, Min, Ed. D.
This study examined how a multimedia-based cognitive tool (MCT) facilitates novices' conceptual understanding of object-oriented programming (OOP). The tool used in this study was CSNüb, a game template created in Adobe Flash. The MCT design framework guided CSNüb's design. The MCT design framework was synthesized from literature on constructivist, multimedia, and motivation learning theories and computer-based cognitive tool design principles. Students worked with CSNüb to develop a simple role-playing game (RPG). Through clinical interviews and process tracing methods, it was found that CSNüb affected novice computer science students' conceptual understanding of OOP through five cognitive processes and factors: cognitive disequilibrium evoked through multimedia-based feedback, exploring for resources that scaffold understanding, changing the level of awareness of the "bigger picture" and ability for higher-level thinking, and consistent refinement of solutions and mental models within the problem space. The five cognitive processes and factors were found to be the result of three levels of interaction with CSNüb. At the Tool Level, students received conflicting information, generally through multimedia-based feedback from the CSNüb, which placed students in states of disequilibrium. At the Interaction Level, students interacted with the CSNüb to resolve their disequilibrium through exploring resources within the tool and refining their solution. They were able to experiment and test out their understanding on OOP. At the Cognitive Level, students used the resources as cognitive scaffolds found through exploration, which in turn, increased the degree of awareness and influenced the level at which they understood the object-oriented system. The five cognitive processes and factors through the three levels of interaction were formed into one model--the MCT Interaction Model (MCTIM)--as a general explanation for how MCTs, such as CSNüb, affects novice students' conceptual understanding.