Believable and Effective AI Agents in Virtual Worlds

Umarov, Iskander; Mozgovoy, Maxim

doi:10.4018/jgcms.2012040103

Cited by 25 publications

(6 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Believability is also considered a performance target in agent-based systems designed by Riedl and Young (2005), Gorman et al (2006), Bogdanovych et al (2016), andDogan (2016). In another approach, Raj and Balakrishnan (2011), Umarov et al (2012), Bevacqua et al (2017), andSutoyo et al (2019) have tested believability as a KPI performance of AI agents design. Mezzanzanica et al (2015) and Karuna et al (2021) discover that believability consists of corrective interventions, thus simplifying the process of developing data cleaning activities.…”

Section: Believabilitymentioning

confidence: 99%

Understanding Entrepreneurial Opportunity Evaluations: An Institutional Logics Approach

Garud,

Pati,

Sutter

et al. 2024

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

Ownership information of high value assets such as property is often concealed and fragmented, adversely affecting information believability. Following the design science research approach, we conceptualize believability as a data quality dimension that supports ownership traceability. We then investigate how blockchain technology might improve information believability in ownership traceability systems. We represent and address our findings via the development of a blockchain-based continuous timestamps tracking system model, framework and implementation for property ownership. A use case of banking transactional data for property ownership traceability is introduced to illustrate our workflow and system design. The proposed system takes advantage of blockchain technology such as traceability and irreversibility to support information believability in the design, management, and use of information systems.

show abstract

Section: Believabilitymentioning

confidence: 99%

Understanding Entrepreneurial Opportunity Evaluations: An Institutional Logics Approach

Garud,

Pati,

Sutter

et al. 2024

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

show abstract

“…However, a more common approach to measuring believability is through subjective assessment [1], [5], [13]- [17] where, similarly to the traditional Turing Test [18], human subjects are asked to observe a character in a game and indicate whether they believe it is being controlled by a human or by a computer. In this paper we follow the subjective assessment approach and we crowdsource rank annotations of believability given to video recorded player characters of a Super Mario Bros variant.…”

Section: A Believability Believable Agents and Their Assessmentmentioning

confidence: 99%

Platformer level design for player believability

Camilleri

Yannakakis

Dingli

2016

2016 IEEE Conference on Computational Intelligence and Games (CIG)

View full text Add to dashboard Cite

Abstract-Player believability is often defined as the ability of a game playing character to convince an observer that it is being controlled by a human. The agent's behavior is often assumed to be the main contributor to the character's believability. In this paper we reframe this core assumption and instead focus on the impact of the game environment and aspects of game design (such as level design) on the believability of the game character. To investigate the relationship between game content and believability we crowdsource rank-based annotations from subjects that view playthrough videos of various AI and human controlled agents in platformer levels of dissimilar characteristics. For this initial study we use a variant of the well-known Super Mario Bros game. We build support vector machine models of reported believability based on gameplay and level features which are extracted from the videos. The highest performing model predicts perceived player believability of a character with an accuracy of 73.31%, on average, and implies a direct relationship between level features and player believability.

show abstract

“…Roughly, we could distinguish 2 goals in video game AI research. Some work aims at creating agents with properties that make them more fun to play with such as human-like behavior [1,2]. Competitions like BotPrize or the Turing test track of the Mario AI Championship focus on this goal.…”

Section: Related Workmentioning

confidence: 99%

Imitative learning for real-time strategy games

Gemine

Safadi

Fonteneau

et al. 2012

2012 IEEE Conference on Computational Intelligence and Games (CIG)

View full text Add to dashboard Cite

Over the past decades, video games have become increasingly popular and complex. Virtual worlds have gone a long way since the first arcades and so have the artificial intelligence (AI) techniques used to control agents in these growing environments. Tasks such as world exploration, constrained pathfinding or team tactics and coordination just to name a few are now default requirements for contemporary video games. However, despite its recent advances, video game AI still lacks the ability to learn. In this paper, we attempt to break the barrier between video game AI and machine learning and propose a generic method allowing real-time strategy (RTS) agents to learn production strategies from a set of recorded games using supervised learning. We test this imitative learning approach on the popular RTS title StarCraft II R and successfully teach a Terran agent facing a Protoss opponent new production strategies.

show abstract

Believable and Effective AI Agents in Virtual Worlds

Cited by 25 publications

References 45 publications

Understanding Entrepreneurial Opportunity Evaluations: An Institutional Logics Approach

Understanding Entrepreneurial Opportunity Evaluations: An Institutional Logics Approach

Platformer level design for player believability

Imitative learning for real-time strategy games

Contact Info

Product

Resources

About