Pub Date : 2022-10-11DOI: 10.1609/aiide.v18i1.21979
Mira Fisher
Low-level game environments and other simulations present a difficulty of scale for an expensive AI technique like narrative planning, which is normally constrained to environments with small state spaces. Due to this limitation, the intentional and cooperative behavior of agents guided by this technology cannot be deployed for different systems without significant additional authoring effort. I propose a process for automatically creating models for larger-scale domains such that a narrative planner can be employed in these settings. By generating an abstract domain of an environment while retaining the information needed to produce behavior appropriate to the abstract actions, agents are able to reason in a lower-complexity space and act in the higher-complexity one. This abstraction is accomplished by the development of extended-duration actions and the identification of their preconditions and effects. Together these components may be combined to form a narrative planning domain, and plans from this domain can be executed within the low-level environment.
{"title":"Narrative Planning in Large Domains through State Abstraction and Option Discovery","authors":"Mira Fisher","doi":"10.1609/aiide.v18i1.21979","DOIUrl":"https://doi.org/10.1609/aiide.v18i1.21979","url":null,"abstract":"Low-level game environments and other simulations present a difficulty of scale for an expensive AI technique like narrative planning, which is normally constrained to environments with small state spaces. Due to this limitation, the intentional and cooperative behavior of agents guided by this technology cannot be deployed for different systems without significant additional authoring effort. I propose a process for automatically creating models for larger-scale domains such that a narrative planner can be employed in these settings. By generating an abstract domain of an environment while retaining the information needed to produce behavior appropriate to the abstract actions, agents are able to reason in a lower-complexity space and act in the higher-complexity one. This abstraction is accomplished by the development of extended-duration actions and the identification of their preconditions and effects. Together these components may be combined to form a narrative planning domain, and plans from this domain can be executed within the low-level environment.","PeriodicalId":92576,"journal":{"name":"Proceedings. AAAI Artificial Intelligence and Interactive Digital Entertainment Conference","volume":"157 1","pages":"299-302"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77437093","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-11DOI: 10.1609/aiide.v18i1.21970
Nick Junius, Michael Mateas, Noah Wardrip-Fruin, Elín Carstensdóttir
Interactive drama has focused on how to allow the player agency in influencing their experience through plot action. Interactive drama's preoccupation with changing plot structure bears little resemblance to theater's emphasis on character expression and dramatic play. Dramatic play allows the player to embody the character through actions, focusing on how characters express themselves and react rather than on how or even if that impacts the overarching sequence of events. In this paper, we present Puppitor, a system for character expression of emotion for interactive storytelling built using acting practice and fighting games as the foundation for its core design and describe its usage in conjunction with Ren'Py to facilitate a novel interactive narrative experience as a case study.
{"title":"Playing with the Strings: Designing Puppitor as an Acting Interface for Digital Games","authors":"Nick Junius, Michael Mateas, Noah Wardrip-Fruin, Elín Carstensdóttir","doi":"10.1609/aiide.v18i1.21970","DOIUrl":"https://doi.org/10.1609/aiide.v18i1.21970","url":null,"abstract":"Interactive drama has focused on how to allow the player agency in influencing their experience through plot action. Interactive drama's preoccupation with changing plot structure bears little resemblance to theater's emphasis on character expression and dramatic play. Dramatic play allows the player to embody the character through actions, focusing on how characters express themselves and react rather than on how or even if that impacts the overarching sequence of events. In this paper, we present Puppitor, a system for character expression of emotion for interactive storytelling built using acting practice and fighting games as the foundation for its core design and describe its usage in conjunction with Ren'Py to facilitate a novel interactive narrative experience as a case study.","PeriodicalId":92576,"journal":{"name":"Proceedings. AAAI Artificial Intelligence and Interactive Digital Entertainment Conference","volume":"19 1","pages":"250-257"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76741011","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-11DOI: 10.1609/aiide.v18i1.21971
A. Liu
Mechanics involving the roll of multiple dice---a "dice pool"---commonly appear in tabletop board games and role-playing games. Existing general-purpose dice pool probability calculators resort to exhaustive enumeration of all possible sorted sequences of rolls, which can quickly become computationally intractable. We propose a dynamic programming algorithm that can efficiently compute probabilities for a wide variety of dice pool mechanics while limiting the need for bespoke optimization. We also present Icepool, a pure Python implementation of the algorithm combined with a library of common dice operations.
{"title":"Icepool: Efficient Computation of Dice Pool Probabilities","authors":"A. Liu","doi":"10.1609/aiide.v18i1.21971","DOIUrl":"https://doi.org/10.1609/aiide.v18i1.21971","url":null,"abstract":"Mechanics involving the roll of multiple dice---a \"dice pool\"---commonly appear in tabletop board games and role-playing games. Existing general-purpose dice pool probability calculators resort to exhaustive enumeration of all possible sorted sequences of rolls, which can quickly become computationally intractable. We propose a dynamic programming algorithm that can efficiently compute probabilities for a wide variety of dice pool mechanics while limiting the need for bespoke optimization. We also present Icepool, a pure Python implementation of the algorithm combined with a library of common dice operations.","PeriodicalId":92576,"journal":{"name":"Proceedings. AAAI Artificial Intelligence and Interactive Digital Entertainment Conference","volume":"22 1","pages":"258-265"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83471699","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-11DOI: 10.1609/aiide.v18i1.21942
R. E. Cardona-Rivera, J. Zagal, Michael S. Debus
Game system models introduce abstractions over games in order to support their analysis, generation, and design. While excellent, models to date leave tacit what they abstract over, why they are ontologically adequate, and how they would be realized in the engine underlying the game. In this paper we model these abstraction gaps via the first-order modal mu-calculus. We use it to reify the link between engines to our game interaction model, a player-computer interaction framework grounded in the Game Ontology Project. Through formal derivation and justification, we contend our work is a useful code studies perspective that affords better understanding the semantics underlying game system models in general.
{"title":"Game System Models: Toward Semantic Foundations for Technical Game Analysis, Generation, and Design","authors":"R. E. Cardona-Rivera, J. Zagal, Michael S. Debus","doi":"10.1609/aiide.v18i1.21942","DOIUrl":"https://doi.org/10.1609/aiide.v18i1.21942","url":null,"abstract":"Game system models introduce abstractions over games in order to support their analysis, generation, and design. While excellent, models to date leave tacit what they abstract over, why they are ontologically adequate, and how they would be realized in the engine underlying the game. In this paper we model these abstraction gaps via the first-order modal mu-calculus. We use it to reify the link between engines to our game interaction model, a player-computer interaction framework grounded in the Game Ontology Project. Through formal derivation and justification, we contend our work is a useful code studies perspective that affords better understanding the semantics underlying game system models in general.","PeriodicalId":92576,"journal":{"name":"Proceedings. AAAI Artificial Intelligence and Interactive Digital Entertainment Conference","volume":"52 1","pages":"10-17"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79555610","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-11DOI: 10.1609/aiide.v18i1.21953
A. Goslen, Daniel Carpenter, Jonathan P. Rowe, R. Azevedo, James C. Lester
Plan recognition is a key component of player modeling. Player plan recognition focuses on modeling how and when players select goals and formulate action sequences to achieve their goals during gameplay. By occasionally asking players to describe their plans, it is possible to devise robust plan recognition models that jointly reason about player goals and action sequences in coordination with player input. In this work, we present a player plan recognition framework that leverages data from player interactions with a planning support tool embedded in an educational game for middle school science education, CRYSTAL ISLAND. Players are prompted to use the planning tool to describe their goals and planned actions in CRYSTAL ISLAND. We use this data to devise data-driven player plan recognition models using multi-label multi-task learning. Specifically, we compare single-task and multi-task learning approaches for both goal prediction and action sequence prediction. Results indicate that multi-task learning yields significant benefits for action sequence prediction. Additionally, we find that incorporating automated detectors of plan completion in plan recognition models improves predictive performance in both tasks.
{"title":"Robust Player Plan Recognition in Digital Games with Multi-Task Multi-Label Learning","authors":"A. Goslen, Daniel Carpenter, Jonathan P. Rowe, R. Azevedo, James C. Lester","doi":"10.1609/aiide.v18i1.21953","DOIUrl":"https://doi.org/10.1609/aiide.v18i1.21953","url":null,"abstract":"Plan recognition is a key component of player modeling. Player plan recognition focuses on modeling how and when players select goals and formulate action sequences to achieve their goals during gameplay. By occasionally asking players to describe their plans, it is possible to devise robust plan recognition models that jointly reason about player goals and action sequences in coordination with player input. In this work, we present a player plan recognition framework that leverages data from player interactions with a planning support tool embedded in an educational game for middle school science education, CRYSTAL ISLAND. Players are prompted to use the planning tool to describe their goals and planned actions in CRYSTAL ISLAND. We use this data to devise data-driven player plan recognition models using multi-label multi-task learning. Specifically, we compare single-task and multi-task learning approaches for both goal prediction and action sequence prediction. Results indicate that multi-task learning yields significant benefits for action sequence prediction. Additionally, we find that incorporating automated detectors of plan completion in plan recognition models improves predictive performance in both tasks.","PeriodicalId":92576,"journal":{"name":"Proceedings. AAAI Artificial Intelligence and Interactive Digital Entertainment Conference","volume":"40 1","pages":"105-112"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87794405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-11DOI: 10.1609/aiide.v18i1.21962
Rushit Sanghrajka, R. Young
A growing number of algorithms for story planning include the ability to create stories with failed actions -- in particular failed actions that occur because of the mistaken beliefs of the characters attempting them. To date, most of these systems have been evaluated analytically, primarily by comparing their expressive range to prior story generation systems. Empirical evaluation of these systems has been preliminary. In this paper, we outline a general comprehension-based approach to the evaluation of plan-based story generation. We describe how we specialize it for use evaluating story plans containing failed actions, and we describe the design and results of an experiment using this approach to evaluate plot lines produced by HeadSpace, a system that models the beliefs of characters and uses that model to generate plot lines containing actions that are attempted but that fail.
{"title":"Evaluating Reader Comprehension of Plan-Based Stories Containing Failed Actions","authors":"Rushit Sanghrajka, R. Young","doi":"10.1609/aiide.v18i1.21962","DOIUrl":"https://doi.org/10.1609/aiide.v18i1.21962","url":null,"abstract":"A growing number of algorithms for story planning include the ability to create stories with failed actions -- in particular failed actions that occur because of the mistaken beliefs of the characters attempting them. To date, most of these systems have been evaluated analytically, primarily by comparing their expressive range to prior story generation systems. Empirical evaluation of these systems has been preliminary. In this paper, we outline a general comprehension-based approach to the evaluation of plan-based story generation. We describe how we specialize it for use evaluating story plans containing failed actions, and we describe the design and results of an experiment using this approach to evaluate plot lines produced by HeadSpace, a system that models the beliefs of characters and uses that model to generate plot lines containing actions that are attempted but that fail.","PeriodicalId":92576,"journal":{"name":"Proceedings. AAAI Artificial Intelligence and Interactive Digital Entertainment Conference","volume":"93 1","pages":"179-188"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88933905","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-11DOI: 10.1609/aiide.v18i1.21972
G. Mori, D. Thue, Stephan Schiffel
Experience Management uses AI technologies to improve people's experiences within an interactive application by changing the environment while the experience is underway. Game-related research in this field has a trend where each experience manager is built in a way that is tightly integrated with the environment that it can change. One consequence of this integration is that it becomes difficult to compare one manager to another in a single environment, or a single manager to itself across multiple environments. With this paper, we propose a solution for decoupling experience managers from the environments that they can change, through the use of an intermediate software platform. We describe the structure of the platform, a protocol that facilitates communication between a manager and an environment, and how normal communication happens. Moreover, we introduce the Camelot Wrapper, software built to extend the interactive visualization engine Camelot and connect it to our platform.
{"title":"EM-Glue: A Platform for Decoupling Experience Managers and Environments","authors":"G. Mori, D. Thue, Stephan Schiffel","doi":"10.1609/aiide.v18i1.21972","DOIUrl":"https://doi.org/10.1609/aiide.v18i1.21972","url":null,"abstract":"Experience Management uses AI technologies to improve people's experiences within an interactive application by changing the environment while the experience is underway. Game-related research in this field has a trend where each experience manager is built in a way that is tightly integrated with the environment that it can change. One consequence of this integration is that it becomes difficult to compare one manager to another in a single environment, or a single manager to itself across multiple environments. With this paper, we propose a solution for decoupling experience managers from the environments that they can change, through the use of an intermediate software platform. We describe the structure of the platform, a protocol that facilitates communication between a manager and an environment, and how normal communication happens. Moreover, we introduce the Camelot Wrapper, software built to extend the interactive visualization engine Camelot and connect it to our platform.","PeriodicalId":92576,"journal":{"name":"Proceedings. AAAI Artificial Intelligence and Interactive Digital Entertainment Conference","volume":"173 1","pages":"266-274"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84293677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-11DOI: 10.1609/aiide.v18i1.21968
Michael Cook
In this paper we introduce Puck, a new automated game design system which combines continuous creativity with an exhaustive approach to content generation. We explain the motivation behind Puck, and in particular its focus on users and small communities. Puck is, to our knowledge, the first automated game designer that can be downloaded and individualise itself through testing and design. We then describe the engineering and structure of the system, detail some initial outputs and evaluation of the system, and future work.
{"title":"Puck: A Slow and Personal Automated Game Designer","authors":"Michael Cook","doi":"10.1609/aiide.v18i1.21968","DOIUrl":"https://doi.org/10.1609/aiide.v18i1.21968","url":null,"abstract":"In this paper we introduce Puck, a new automated game design system which combines continuous creativity with an exhaustive approach to content generation. We explain the motivation behind Puck, and in particular its focus on users and small communities. Puck is, to our knowledge, the first automated game designer that can be downloaded and individualise itself through testing and design. We then describe the engineering and structure of the system, detail some initial outputs and evaluation of the system, and future work.","PeriodicalId":92576,"journal":{"name":"Proceedings. AAAI Artificial Intelligence and Interactive Digital Entertainment Conference","volume":"47 1","pages":"232-239"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90393226","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-11DOI: 10.1609/aiide.v18i1.21956
Eric W. Lang, R. Young
Recent work extending planning algorithms that reason about action and change has been successful at supporting game design, player modeling, and story generation. Incorporating agent preferences over actions and propositions into a planning process allows for a more accurate prediction of what a human might do when solving a problem like playing through a game level. This paper presents the preference-based planning heuristic RPGPref which uses relaxed plan graphs (RPGs) and preference sets to guide a planner toward a preference-conforming path to its goal. A human subjects evaluation confirms that RPGPref successfully guides the planning process toward solution plans that recognizably match and differentiate player playstyles.
{"title":"RPGPref: A Planning Heuristic That Uses Playstyle Preferences to Model Player Action and Choice","authors":"Eric W. Lang, R. Young","doi":"10.1609/aiide.v18i1.21956","DOIUrl":"https://doi.org/10.1609/aiide.v18i1.21956","url":null,"abstract":"Recent work extending planning algorithms that reason about action and change has been successful at supporting game design, player modeling, and story generation. Incorporating agent preferences over actions and propositions into a planning process allows for a more accurate prediction of what a human might do when solving a problem like playing through a game level. This paper presents the preference-based planning heuristic RPGPref which uses relaxed plan graphs (RPGs) and preference sets to guide a planner toward a preference-conforming path to its goal. A human subjects evaluation confirms that RPGPref successfully guides the planning process toward solution plans that recognizably match and differentiate player playstyles.","PeriodicalId":92576,"journal":{"name":"Proceedings. AAAI Artificial Intelligence and Interactive Digital Entertainment Conference","volume":"8 1","pages":"129-136"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90383204","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-11DOI: 10.1609/aiide.v18i1.21977
Kristen K. Yu, Matthew J. Guzdial, Nathan R Sturtevant
In AI director research, it is not straightforward for researchers to understand how each algorithm affects the player experience. This demo introduces PWR, which is a new fully developed video game test bed to evaluate AI directors. This demo includes 3 different AI director algorithms in order to help researchers improve their intuition for understanding the differences between potential algorithms, and also provides insight on the framework required to author a new AI director. This test bed can support future AI director research by allowing for direct comparisons of new algorithms.
{"title":"FarmQuest: A Demonstration of an AI Director Video Game Test Bed","authors":"Kristen K. Yu, Matthew J. Guzdial, Nathan R Sturtevant","doi":"10.1609/aiide.v18i1.21977","DOIUrl":"https://doi.org/10.1609/aiide.v18i1.21977","url":null,"abstract":"In AI director research, it is not straightforward for researchers to understand how each algorithm affects the player experience. This demo introduces PWR, which is a new fully developed video game test bed to evaluate AI directors. This demo includes 3 different AI director algorithms in order to help researchers improve their intuition for understanding the differences between potential algorithms, and also provides insight on the framework required to author a new AI director. This test bed can support future AI director research by allowing for direct comparisons of new algorithms.","PeriodicalId":92576,"journal":{"name":"Proceedings. AAAI Artificial Intelligence and Interactive Digital Entertainment Conference","volume":"55 1","pages":"288-290"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76907578","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：