IEEE Transactions on Games最新文献

{"title":"Goal-Oriented Interactions in Games Using LLMs","authors":"Adon Phillips;Jochen Lang;David Mould","doi":"10.1109/TG.2024.3515807","DOIUrl":"https://doi.org/10.1109/TG.2024.3515807","url":null,"abstract":"Unrealistic parser-based dialogue systems limit player agency. Large language model (LLM) characters can enhance agency but lack structure and measurable objectives. In this article, we propose a framework for structured interactions that tracks player progress through specific objectives, while also improving character LLM responses. This approach frames interactions as puzzles with states representing goal-based milestones. We employ an LLM to analyze dialogue history and enforce state transitions for state awareness and to enable specific actions like tailored LLM prompts and multimodal content changes. This results in a robust dialogue state tracking system for goal-based interactions. Using our method, a designer can craft transition rules as abstract goals that allow players to invent their own solutions rather than discovering the designer's intent. We demonstrate this with a hostage scenario game, where the player negotiates with a hostage-taker adversary. The game's effectiveness is assessed through qualitative gameplay analysis and a quantitative evaluation of our state tracking method.","PeriodicalId":55977,"journal":{"name":"IEEE Transactions on Games","volume":"17 2","pages":"510-521"},"PeriodicalIF":1.7,"publicationDate":"2024-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308426","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Research on the Imperfect Information Game of Four-Player Mahjong Based on Mix-PPO","authors":"Jia-Yang Wang;Ming-Yan Wang;Wang Zeng;Zi-An Zhong","doi":"10.1109/TG.2024.3507107","DOIUrl":"https://doi.org/10.1109/TG.2024.3507107","url":null,"abstract":"In recent years, the deep reinforcement learning method has performed well in many challenging tasks, including <italic>Go</i> and <italic>MOBA</i> games and other industrial fields. <italic>Mahjong</i> is a popular game with imperfect information, but because of its large amount of hidden information and complex game rules, it is very challenging to solve its game intelligence decision problem and build artificial intelligence beyond the human level. To solve the aforementioned problems, this article proposes a feature encoding method and model training strategy for four players in Chinese <italic>Mahjong</i>. In addition, this article also innovatively proposes the Mix-PPO algorithm, which combines the advantages of the traditional PPO1 algorithm and the PPO2 algorithm, and compares the Mix-PPO algorithm with other algorithms, including the traditional proximal policy optimization algorithm, the deep-learning-related algorithm, and the game search tree algorithm. The experimental results show the validity of the feature coding and the Mix-PPO algorithm of Chinese four-player <italic>Mahjong</i> in building the decision-making model of Chinese four-player <italic>Mahjong</i>, as well as the validity of the coding method of the model's <italic>Mahjong</i> feature and the training strategy.","PeriodicalId":55977,"journal":{"name":"IEEE Transactions on Games","volume":"17 2","pages":"485-497"},"PeriodicalIF":1.7,"publicationDate":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10776753","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Know Your Game, From in-Real Life Experts to Video Game Experts: Discriminating in-Real Life Experts From Non Experts Using Blinks and EAR-Derived Features","authors":"Gianluca Guglielmo;Michal Klincewicz;Elisabeth Huis in‘t Veld;Pieter Spronck","doi":"10.1109/TG.2024.3494724","DOIUrl":"https://doi.org/10.1109/TG.2024.3494724","url":null,"abstract":"Serious games are an effective method of reproducing aspects of the complex interplay between environments and stakeholders in business situations. In the game, we describe here, <italic>The Sustainable Port</i>, players experience what it is like to make decisions in such a complex environment. Their aim in the game is to grow the Port of Rotterdam while keeping economic growth in balance with sustainability goals. In this study, we assessed whether experienced Port of Rotterdam employees (PoR employees) show different psychophysiological patterns, and more specifically eye aspect ratio (EAR)-derived features, compared to students. We did this on the assumption that physiological patterns will tell us something about how people who are familiar with the environment of the Port of Rotterdam, more specifically PoR employees, make decisions compared to those lacking such familiarity. Our sample consisted of 28 PoR employees and 65 students, all of whom played <italic>The Sustainable Port</i> game and had their faces recorded with a camera. The EAR was extracted from these recordings, and then from those, we extracted EAR-derived features. Our results show that PoR employees perform better than students and that the two groups are characterized by different physiological variations in their EAR-derived features. A logistic regression model used to identify PoR employees and students obtained an F1 score of 0.62, an area under the precision–recall curve score of 0.64, and an ROC AUC score of 0.70. Such a performance significantly above baseline suggests the effectiveness of using EAR-derived features for this task. Our interpretation was further confirmed by a pseudo-R2 score used to evaluate the goodness of fit of a logistic regression model on the entire dataset. We found that PoR employees had a lower variation in blink rate per minute and higher variation in the root mean square of the successive differences in blinks (RMSSD), the consecutive difference between two continuous blinks. Moreover, this study shows that our methods were robust enough to negate the effects of confounders, such as biological sex and age, that affect some other studies that analyze blinks.","PeriodicalId":55977,"journal":{"name":"IEEE Transactions on Games","volume":"17 2","pages":"474-484"},"PeriodicalIF":1.7,"publicationDate":"2024-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"User-Centric Locomotion Techniques for Virtual Reality Games: A Survey of User Needs and Issues","authors":"Daichi Hirobe;Shizuka Shirai;Jason Orlosky;Mehrasa Alizadeh;Masato Kobayashi;Yuki Uranishi;Photchara Ratsamee;Haruo Takemura","doi":"10.1109/TG.2024.3498325","DOIUrl":"https://doi.org/10.1109/TG.2024.3498325","url":null,"abstract":"Virtual reality (VR) video games that are played on a VR headset are becoming increasingly common in households, and though many games require players to navigate vast virtual spaces, most homes cannot provide a large enough physical space to encompass the entire virtual space. Thus, VR video games that require locomotion often provide users with alternative locomotion techniques. While teleportation or steering is typically used as a standard, new techniques can overcome remaining problems, such as motion sickness. However, a holistic perspective of user needs and issues regarding these techniques in practical situations has not been studied on a broad basis. To address this gap in the literature and contribute to future VR video game development and research, we conducted 16 semi-structured interviews and surveyed 88 participants to help explore issues regarding existing locomotion techniques. Our results revealed preferences related to teleportation versus steering and the postures that users adopt while playing VR video games, along with user needs for locomotion techniques in each posture.","PeriodicalId":55977,"journal":{"name":"IEEE Transactions on Games","volume":"17 2","pages":"460-473"},"PeriodicalIF":1.7,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10753078","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308263","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Will GPT-4 Run DOOM?","authors":"Adrian de Wynter","doi":"10.1109/TG.2024.3497601","DOIUrl":"https://doi.org/10.1109/TG.2024.3497601","url":null,"abstract":"We show that GPT-4’s reasoning and planning capabilities extend to the 1993 first-person shooter <italic>Doom</i>. This large language model (LLM) is able to run and play the game with only a few instructions, plus a textual description–generated by the model itself from screenshots–about the state of the game being observed. We find that GPT-4 can play the game to a passable degree: it is able to manipulate doors, combat enemies, and perform pathing. More complex prompting strategies involving multiple model calls provide better results. While further work is required to enable the LLM to play the game as well as its classical, reinforcement learning-based counterparts, we note that GPT-4 required no training, leaning instead on its own reasoning and observational capabilities. We hope our work pushes the boundaries on intelligent, LLM-based agents in video games. We conclude by discussing the ethical implications of our work.","PeriodicalId":55977,"journal":{"name":"IEEE Transactions on Games","volume":"17 2","pages":"451-459"},"PeriodicalIF":1.7,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10752360","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308423","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Implicit Coordination Dynamics: A Synchrony-Based Study on Team Positioning and Performance in Competitive Dota 2","authors":"Udesh Habaraduwa;Paris Mavromoustakos Blom;Travis J. Wiltshire","doi":"10.1109/TG.2024.3488114","DOIUrl":"https://doi.org/10.1109/TG.2024.3488114","url":null,"abstract":"The collective performance displayed by groups or teams, whether in solving complex problems or excelling in esports competitions, hinges on their coordination dynamics. While explicit coordination (e.g., verbal commands) and its affects on collective outcomes have been studied extensively, implicit coordination, especially in dynamic, fast-paced environments have been under investigated. In this study, we examine the competitive esport <italic>Dota 2</i> (D2) as a setting to explore within-team implicit coordination, which we model as player avatar movement synchrony. Utilizing the cluster phase method, we analyze spatio-temporal patterns of player movements to quantify implicit movement coordination. We observe a negative linear relationship between team movement synchrony and team performance in rank for D2 competitions across two tournaments. While some research suggests stronger coordination leads to favorable outcomes, we leverage our findings to discuss the complexity of team coordination, showcasing a delicate balance between specialization of individual team members and collective action. This study not only extends complex systems techniques used in physical sports to the rapidly evolving esports arena, but also invites further exploration into the multidimensional nature of coordination in team-based activities.","PeriodicalId":55977,"journal":{"name":"IEEE Transactions on Games","volume":"17 2","pages":"442-450"},"PeriodicalIF":1.7,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308422","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Genetic Algorithm for Solving Sudoku Based on Multiarmed Bandit Selection","authors":"Jon-Lark Kim;Eunjee Eor","doi":"10.1109/TG.2024.3487861","DOIUrl":"https://doi.org/10.1109/TG.2024.3487861","url":null,"abstract":"In this article, we introduce a genetic algorithm-based upper confidence bound (GA-UCB), an innovative hybrid genetic algorithm integrating multiarmed bandit. It effectively addresses the challenges of solving large and intricate <italic>Sudoku</i> puzzles, thus overcoming the constraints of traditional genetic algorithms. In GA-UCB, reinforcement learning is applied to simulate parent selection and crossover. By learning the optimal parent selection within a given population, the population evolves. Based on this technology, GA-UCB demonstrates improved results in solving complex <italic>Sudoku</i> puzzles. GA-UCB is compared with several state-of-the-art algorithms on <italic>Sudoku</i> puzzles of different difficulty levels and shows a 55% improvement in convergence speed compared to previous research results, particularly in the most challenging instance among the six <italic>Sudoku</i> puzzle instances tested.","PeriodicalId":55977,"journal":{"name":"IEEE Transactions on Games","volume":"17 2","pages":"429-441"},"PeriodicalIF":1.7,"publicationDate":"2024-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308215","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Collaborative AI in the Geometry Friends Game Competition","authors":"Alberto Almagro;Juan Carlos Llamas-Núñez;Antonio A. Sánchez-Ruiz;Belén Díaz-Agudo","doi":"10.1109/TG.2024.3487309","DOIUrl":"https://doi.org/10.1109/TG.2024.3487309","url":null,"abstract":"<italic>Geometry Friends</i> is a collaborative platform game featuring two agents, a circle and a rectangle, navigating through physics-driven levels to collect diamonds. Each level presents challenges in the form of obstacles and diamonds arranged in different configurations that forces the agents to cooperate. In this article, we describe the Universidad Complutense de Madrid (UCM) agents, that participated in the 2023 <italic>Geometry Friends</i> AI Competition with excellent results, especially in the collaborative track. Our approach combines level analysis, motion simulation, abstract planning, scripted actions, reinforcement learning, and different types of agent interaction. The contributions are versatile and potentially applicable to other collaborative platform games.","PeriodicalId":55977,"journal":{"name":"IEEE Transactions on Games","volume":"17 2","pages":"408-418"},"PeriodicalIF":1.7,"publicationDate":"2024-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10737403","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308425","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Large Language Models as Narrative Planning Search Guides","authors":"Rachelyn Farrell;Stephen G. Ware","doi":"10.1109/TG.2024.3487416","DOIUrl":"https://doi.org/10.1109/TG.2024.3487416","url":null,"abstract":"Symbolic planning algorithms and large language models have different strengths and weaknesses for story generation, suggesting hybrid models might leverage advantages from both. Others have proposed using a language model in combination with a partial order planning style algorithm to avoid the need for a hand-written symbolic domain of actions, or generating these domains from natural language input. This article offers a complementary approach. We propose to use a state space planning algorithm to plan coherent multiagent stories using hand-written symbolic domains, but with a language model acting as a guide to estimate, which events are worth exploring first. We present an initial evaluation of this approach on a set of benchmark narrative planning problems.","PeriodicalId":55977,"journal":{"name":"IEEE Transactions on Games","volume":"17 2","pages":"419-428"},"PeriodicalIF":1.7,"publicationDate":"2024-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308482","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CuDA2: An Approach for Incorporating Traitor Agents Into Cooperative Multiagent Systems","authors":"Zhen Chen;Yong Liao;Youpeng Zhao;Zipeng Dai;Jian Zhao","doi":"10.1109/TG.2024.3485726","DOIUrl":"https://doi.org/10.1109/TG.2024.3485726","url":null,"abstract":"Cooperative multiagent reinforcement learning (CMARL) strategies are well known to be vulnerable to adversarial perturbations. Previous works on adversarial attacks have primarily focused on glass-box attacks that directly perturb the states or actions of victim agents, often in scenarios with a limited number of attacks. However, gaining complete access to victim agents in real-world environments is exceedingly difficult. To create more realistic adversarial attacks, we introduce a novel method that involves injecting traitor agents into the CMARL system. We model this problem as a traitor Markov decision process (TMDP), where traitors cannot directly attack the victim agents but can influence their formation or positioning through collisions. In TMDP, traitors are trained using the same MARL algorithm as the victim agents, with their reward function set as the negative of the victim agents' reward. Despite this, the training efficiency for traitors remains low because it is challenging for them to directly associate their actions with the victim agents' rewards. To address this issue, we propose the curiosity-driven adversarial attack (CuDA2) framework. CuDA2 enhances the efficiency and aggressiveness of attacks on the specified victim agents' policies while maintaining the optimal policy invariance of the traitors. Specifically, we employ a pretrained random network distillation module, where the extra reward generated by the RND module encourages traitors to explore states unencountered by the victim agents. Extensive experiments on various scenarios from SMAC demonstrate that our CuDA2 framework offers comparable or superior adversarial attack capabilities compared to other baselines.","PeriodicalId":55977,"journal":{"name":"IEEE Transactions on Games","volume":"17 2","pages":"397-407"},"PeriodicalIF":1.7,"publicationDate":"2024-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}