This paper investigates the effectiveness of hybrids of learning and evolutionary approaches to find weights and topologies for an artificial neural network (ANN) which is used to evaluate board positions for a two-person zero-sum game, the virus game. Two hybrid approaches: evolutionary RPROP (resilient backpropagation) and evolutionary BP (backpropagation) are described and empirically compared with BP, RPROP, iRPROP (improved RPROP) and evolutionary learning approaches. The results show that evolutionary RPROP and evolutionary BP have significantly better generalisation performance than their constituent learning and evolutionary methods.
{"title":"Hybrid Evolutionary Learning Approaches for The Virus Game","authors":"M. Naveed, P. Cowling, M. A. Hossain","doi":"10.1109/CIG.2007.368098","DOIUrl":"https://doi.org/10.1109/CIG.2007.368098","url":null,"abstract":"This paper investigates the effectiveness of hybrids of learning and evolutionary approaches to find weights and topologies for an artificial neural network (ANN) which is used to evaluate board positions for a two-person zero-sum game, the virus game. Two hybrid approaches: evolutionary RPROP (resilient backpropagation) and evolutionary BP (backpropagation) are described and empirically compared with BP, RPROP, iRPROP (improved RPROP) and evolutionary learning approaches. The results show that evolutionary RPROP and evolutionary BP have significantly better generalisation performance than their constituent learning and evolutionary methods.","PeriodicalId":365269,"journal":{"name":"2007 IEEE Symposium on Computational Intelligence and Games","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114367906","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}
D. Bannach, O. Amft, K. Kunze, E. A. Heinz, G. Tröster, P. Lukowicz
We envision to add context awareness and ambient intelligence to edutainment and computer gaming applications in general. This requires mixed-reality setups and ever-higher levels of immersive human-computer interaction. Here, we focus on the automatic recognition of natural human hand gestures recorded by inexpensive, wearable motion sensors. To study the feasibility of our approach, we chose an educational parking game with 3D graphics that employs motion sensors and hand gestures as its sole game controls. Our implementation prototype is based on Java-3D for the graphics display and on our own CRN Toolbox for sensor integration. It shows very promising results in practice regarding game appeal, player satisfaction, extensibility, ease of interfacing to the sensors, and - last but not least - sufficient accuracy of the real-time gesture recognition to allow for smooth game control. An initial quantitative performance evaluation confirms these notions and provides further support for our setup
{"title":"Waving Real Hand Gestures Recorded by Wearable Motion Sensors to a Virtual Car and Driver in a Mixed-Reality Parking Game","authors":"D. Bannach, O. Amft, K. Kunze, E. A. Heinz, G. Tröster, P. Lukowicz","doi":"10.1109/CIG.2007.368076","DOIUrl":"https://doi.org/10.1109/CIG.2007.368076","url":null,"abstract":"We envision to add context awareness and ambient intelligence to edutainment and computer gaming applications in general. This requires mixed-reality setups and ever-higher levels of immersive human-computer interaction. Here, we focus on the automatic recognition of natural human hand gestures recorded by inexpensive, wearable motion sensors. To study the feasibility of our approach, we chose an educational parking game with 3D graphics that employs motion sensors and hand gestures as its sole game controls. Our implementation prototype is based on Java-3D for the graphics display and on our own CRN Toolbox for sensor integration. It shows very promising results in practice regarding game appeal, player satisfaction, extensibility, ease of interfacing to the sensors, and - last but not least - sufficient accuracy of the real-time gesture recognition to allow for smooth game control. An initial quantitative performance evaluation confirms these notions and provides further support for our setup","PeriodicalId":365269,"journal":{"name":"2007 IEEE Symposium on Computational Intelligence and Games","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129898040","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}
In this paper, we use a simple poker game to investigate Bayesian opponent modeling. Opponents are defined in four distinctive styles, and tactics are developed which defeat each of the respective styles. By analyzing the past actions of each opponent, and comparing to action related probabilities, the most challenging opponent is identified, and the strategy employed is one that aims to counter that player. The opponent modeling player plays well against non-reactive player styles, and also performs well when compared to a player that knows the exact styles of each opponent in advance
{"title":"Bayesian Opponent Modeling in a Simple Poker Environment","authors":"Roderick J. S. Baker, P. Cowling","doi":"10.1109/CIG.2007.368088","DOIUrl":"https://doi.org/10.1109/CIG.2007.368088","url":null,"abstract":"In this paper, we use a simple poker game to investigate Bayesian opponent modeling. Opponents are defined in four distinctive styles, and tactics are developed which defeat each of the respective styles. By analyzing the past actions of each opponent, and comparing to action related probabilities, the most challenging opponent is identified, and the strategy employed is one that aims to counter that player. The opponent modeling player plays well against non-reactive player styles, and also performs well when compared to a player that knows the exact styles of each opponent in advance","PeriodicalId":365269,"journal":{"name":"2007 IEEE Symposium on Computational Intelligence and Games","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131128297","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}
We present a first application of a general approach to learn the behavior of NPCs (and other entities) in a game from observing just the graphical output of the game during game play. This allows some understanding of what a human player might be able to learn during game play. The approach uses object tracking and situation-action pairs with the nearest-neighbor rule. For the game of Pong, we were able to predict the correct behavior of the computer controlled components approximately 9 out of 10 times, even if we keep the usage of knowledge about the game (beyond observing the images) at a minimum
{"title":"Extracting NPC behavior from computer games using computer vision and machine learning techniques","authors":"A. Fink, J. Denzinger, John Aycock","doi":"10.1109/CIG.2007.368075","DOIUrl":"https://doi.org/10.1109/CIG.2007.368075","url":null,"abstract":"We present a first application of a general approach to learn the behavior of NPCs (and other entities) in a game from observing just the graphical output of the game during game play. This allows some understanding of what a human player might be able to learn during game play. The approach uses object tracking and situation-action pairs with the nearest-neighbor rule. For the game of Pong, we were able to predict the correct behavior of the computer controlled components approximately 9 out of 10 times, even if we keep the usage of knowledge about the game (beyond observing the images) at a minimum","PeriodicalId":365269,"journal":{"name":"2007 IEEE Symposium on Computational Intelligence and Games","volume":"136 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133992694","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}
This paper describes the EvoTanks research project, a continuing attempt to develop strong AI players for a primitive `combat' style video game using evolutionary computational methods with artificial neural networks. A small but challenging feat due to the necessity for agent's actions to rely heavily on opponent behaviour. Previous investigation has shown the agents are capable of developing high performance behaviours by evolving against scripted opponents; however these are local to the trained opponent. The focus of this paper shows results from the use of co-evolution on the same population. Results show agents no longer succumb to trappings of local maxima within the search space and are capable of converging on high fitness behaviours local to their population without the use of scripted opponents
{"title":"EvoTanks: Co-Evolutionary Development of Game-Playing Agents","authors":"Thomas Thompson, J. Levine, G. Hayes","doi":"10.1109/CIG.2007.368116","DOIUrl":"https://doi.org/10.1109/CIG.2007.368116","url":null,"abstract":"This paper describes the EvoTanks research project, a continuing attempt to develop strong AI players for a primitive `combat' style video game using evolutionary computational methods with artificial neural networks. A small but challenging feat due to the necessity for agent's actions to rely heavily on opponent behaviour. Previous investigation has shown the agents are capable of developing high performance behaviours by evolving against scripted opponents; however these are local to the trained opponent. The focus of this paper shows results from the use of co-evolution on the same population. Results show agents no longer succumb to trappings of local maxima within the search space and are capable of converging on high fitness behaviours local to their population without the use of scripted opponents","PeriodicalId":365269,"journal":{"name":"2007 IEEE Symposium on Computational Intelligence and Games","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114859687","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}
J. Reisinger, E. Bahçeci, Igor Karpov, R. Miikkulainen
The General Game Playing Competition (Genesereth et al., 2005) poses a unique challenge for artificial intelligence. To be successful, a player must learn to play well in a limited number of example games encoded in first-order logic and then generalize its game play to previously unseen games with entirely different rules. Because good opponents are usually not available, learning algorithms must come up with plausible opponent strategies in order to benchmark performance. One approach to simultaneously learning all player strategies is coevolution. This paper presents a coevolutionary approach using neuroevolution of augmenting topologies to evolve populations of game state evaluators. This approach is tested on a sample of games from the General Game Playing Competition and shown to be effective: It allows the algorithm designer to minimize the amount of domain knowledge built into the system, which leads to more general game play and allows modeling opponent strategies efficiently. Furthermore, the general game playing domain proves to be a powerful tool for developing and testing coevolutionary methods
通用游戏竞赛(Genesereth et al.， 2005)对人工智能提出了一个独特的挑战。要想取得成功，玩家必须学会在有限数量的一阶逻辑游戏中玩得很好，然后将其游戏玩法推广到以前从未见过的、规则完全不同的游戏中。因为通常没有好的对手，所以学习算法必须提出合理的对手策略，以便对性能进行基准测试。同时学习所有玩家策略的一种方法是共同进化。本文提出了一种利用增强拓扑的神经进化来进化博弈状态评估者群体的协同进化方法。这种方法在通用游戏竞赛的游戏样本上进行了测试，并证明是有效的:它允许算法设计者最小化系统中构建的领域知识的数量，从而导致更通用的游戏玩法，并允许有效地建模对手策略。此外，一般博弈域被证明是开发和测试共同进化方法的强大工具
{"title":"Coevolving Strategies for General Game Playing","authors":"J. Reisinger, E. Bahçeci, Igor Karpov, R. Miikkulainen","doi":"10.1109/CIG.2007.368115","DOIUrl":"https://doi.org/10.1109/CIG.2007.368115","url":null,"abstract":"The General Game Playing Competition (Genesereth et al., 2005) poses a unique challenge for artificial intelligence. To be successful, a player must learn to play well in a limited number of example games encoded in first-order logic and then generalize its game play to previously unseen games with entirely different rules. Because good opponents are usually not available, learning algorithms must come up with plausible opponent strategies in order to benchmark performance. One approach to simultaneously learning all player strategies is coevolution. This paper presents a coevolutionary approach using neuroevolution of augmenting topologies to evolve populations of game state evaluators. This approach is tested on a sample of games from the General Game Playing Competition and shown to be effective: It allows the algorithm designer to minimize the amount of domain knowledge built into the system, which leads to more general game play and allows modeling opponent strategies efficiently. Furthermore, the general game playing domain proves to be a powerful tool for developing and testing coevolutionary methods","PeriodicalId":365269,"journal":{"name":"2007 IEEE Symposium on Computational Intelligence and Games","volume":"218 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115524900","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}
Solitaire Yahtzee has been solved completely. However, the optimal strategy is not one a human could practically use, and for computer play it requires either a very large database or significant CPU time. We present some refinements to the techniques used to solve solitaire Yahtzee and give a method for analyzing other solitaire strategies and give some examples of this analysis for some non-optimal strategies, including some produced by evolutionary algorithms
{"title":"Computer Strategies for Solitaire Yahtzee","authors":"James R. Glenn","doi":"10.1109/CIG.2007.368089","DOIUrl":"https://doi.org/10.1109/CIG.2007.368089","url":null,"abstract":"Solitaire Yahtzee has been solved completely. However, the optimal strategy is not one a human could practically use, and for computer play it requires either a very large database or significant CPU time. We present some refinements to the techniques used to solve solitaire Yahtzee and give a method for analyzing other solitaire strategies and give some examples of this analysis for some non-optimal strategies, including some produced by evolutionary algorithms","PeriodicalId":365269,"journal":{"name":"2007 IEEE Symposium on Computational Intelligence and Games","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116106360","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}
Accuracy of evaluation functions is one of the critical factors in computer game players. Evaluation functions are usually constructed manually as a weighted linear combination of evaluation features that characterize game positions. Selecting evaluation features and tuning their weights require deep knowledge of the game and largely alleviates such efforts. In this paper, we propose a new fast and scalable method to automatically generate game position features based on game records to be used in evaluation functions. Our method treats two-class problems which is widely applicable to many types of games. Evaluation features are built as conjunctions of the simplest features representing positions. We select these features based on two measures: frequency and conditional mutual information. To evaluate the proposed method, we applied it to 200,000 Othello positions. The proposed selection method is found to be effective, showing much better results than when simple features are used. The naive Bayesian classifier using automatically generated features showed the accuracy close to 80% in win/lose classification. We also show that this generation method can be parallelized easily and can treat large scale problems by converting these selection algorithms into incremental selection algorithms
{"title":"Automatic Generation of Evaluation Features for Computer Game Players","authors":"Makoto Miwa, Daisaku Yokoyama, T. Chikayama","doi":"10.1109/CIG.2007.368108","DOIUrl":"https://doi.org/10.1109/CIG.2007.368108","url":null,"abstract":"Accuracy of evaluation functions is one of the critical factors in computer game players. Evaluation functions are usually constructed manually as a weighted linear combination of evaluation features that characterize game positions. Selecting evaluation features and tuning their weights require deep knowledge of the game and largely alleviates such efforts. In this paper, we propose a new fast and scalable method to automatically generate game position features based on game records to be used in evaluation functions. Our method treats two-class problems which is widely applicable to many types of games. Evaluation features are built as conjunctions of the simplest features representing positions. We select these features based on two measures: frequency and conditional mutual information. To evaluate the proposed method, we applied it to 200,000 Othello positions. The proposed selection method is found to be effective, showing much better results than when simple features are used. The naive Bayesian classifier using automatically generated features showed the accuracy close to 80% in win/lose classification. We also show that this generation method can be parallelized easily and can treat large scale problems by converting these selection algorithms into incremental selection algorithms","PeriodicalId":365269,"journal":{"name":"2007 IEEE Symposium on Computational Intelligence and Games","volume":"252 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122493728","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}
RoboCupSoccer domain has several leagues which varies in the rule of play such as specification of players, number of players, field size, and time length. Nevertheless, each RoboCup league is a variant of a soccer league and therefore they are based on some basic rules of soccer. A layered design of agents system presented in the work of Garcia et al. (2004) shows a modular approach to build control for a team of robots participating in RoboCupSoccer E-League. Based on this design, we propose a generalized architecture offering flexibility to switch between leagues and programming language while maintaining Prolog as cognitive layer. Prolog is a very convenient tool to design strategies for soccer players using simple rules close to human reasoning. Sometimes this reasoning needs to deal with uncertainty, fuzziness or incompleteness of the information. In these cases it is useful Fuzzy Prolog (Guadarrama et al., 2004), (Munoz-Hernandez and Vaucheret, 2005), (Munoz-Hernandez and Gomez-Perez, 2005), (Munoz-Hernandez and Vaucheret, 2006). In this paper we propose to use a combination of Prolog (that is crisp) and Fuzzy Prolog to implement the cognitive layer in RoboCupSoccer, which has the advantage of incorporating as conventional logic as fuzzy logic in this layer. A prototype of a team based on this architecture has been build for RoboCup soccer simulator, and we show that this approach provides a convenient way of incorporating a team strategy in high level (human-like) manner, where technical details are encapsulated and fuzzy information is represented
RoboCupSoccer领域有几个不同的比赛规则，如球员规格，球员数量，场地大小和时间长度。然而，每个机器人世界杯联赛都是足球联赛的变体，因此它们都基于一些足球的基本规则。Garcia等人(2004)的工作中提出了agent系统的分层设计，展示了一种模块化方法来为参加RoboCupSoccer电子联赛的机器人团队构建控制。基于这种设计，我们提出了一种通用的架构，提供了在联盟和编程语言之间切换的灵活性，同时保持Prolog作为认知层。Prolog是一个非常方便的工具，可以使用接近人类推理的简单规则为足球运动员设计策略。有时这种推理需要处理信息的不确定性、模糊性或不完整性。在这些情况下，Fuzzy Prolog (Guadarrama et al.， 2004)、(Munoz-Hernandez and Vaucheret, 2005)、(Munoz-Hernandez and Gomez-Perez, 2005)、(Munoz-Hernandez and Vaucheret, 2006)是有用的。在本文中，我们建议使用Prolog(即脆)和Fuzzy Prolog的组合来实现RoboCupSoccer中的认知层，其优点是在该层中结合了传统逻辑和模糊逻辑。基于这种架构的团队原型已经为RoboCup足球模拟器构建，我们表明这种方法提供了一种方便的方式，以高水平(类人)的方式合并团队策略，其中技术细节被封装，模糊信息被表示
{"title":"Fuzzy Prolog as Cognitive Layer in RoboCupSoccer","authors":"S. Muñoz-Hernández, W. S. Wiguna","doi":"10.1109/CIG.2007.368118","DOIUrl":"https://doi.org/10.1109/CIG.2007.368118","url":null,"abstract":"RoboCupSoccer domain has several leagues which varies in the rule of play such as specification of players, number of players, field size, and time length. Nevertheless, each RoboCup league is a variant of a soccer league and therefore they are based on some basic rules of soccer. A layered design of agents system presented in the work of Garcia et al. (2004) shows a modular approach to build control for a team of robots participating in RoboCupSoccer E-League. Based on this design, we propose a generalized architecture offering flexibility to switch between leagues and programming language while maintaining Prolog as cognitive layer. Prolog is a very convenient tool to design strategies for soccer players using simple rules close to human reasoning. Sometimes this reasoning needs to deal with uncertainty, fuzziness or incompleteness of the information. In these cases it is useful Fuzzy Prolog (Guadarrama et al., 2004), (Munoz-Hernandez and Vaucheret, 2005), (Munoz-Hernandez and Gomez-Perez, 2005), (Munoz-Hernandez and Vaucheret, 2006). In this paper we propose to use a combination of Prolog (that is crisp) and Fuzzy Prolog to implement the cognitive layer in RoboCupSoccer, which has the advantage of incorporating as conventional logic as fuzzy logic in this layer. A prototype of a team based on this architecture has been build for RoboCup soccer simulator, and we show that this approach provides a convenient way of incorporating a team strategy in high level (human-like) manner, where technical details are encapsulated and fuzzy information is represented","PeriodicalId":365269,"journal":{"name":"2007 IEEE Symposium on Computational Intelligence and Games","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127281872","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}
We investigate the use of genetic algorithms to evolve AI players for real-time strategy games. To overcome the knowledge acquisition bottleneck found in using traditional expert systems, scripts, or decision trees we evolve players through co-evolution. Our game players are implemented as resource allocation systems. Influence map trees are used to analyze the game-state and determine promising places to attack, defend, etc. These spatial objectives are chained to non-spatial objectives (train units, build buildings, gather resources) in a dependency graph. Players are encoded within the individuals of a genetic algorithm and co-evolved against each other, with results showing the production of strategies that are innovative, robust, and capable of defeating a suite of hand-coded opponents
{"title":"Co-Evolving Influence Map Tree Based Strategy Game Players","authors":"C. Miles, J. Quiroz, Ryan E. Leigh, S. Louis","doi":"10.1109/CIG.2007.368083","DOIUrl":"https://doi.org/10.1109/CIG.2007.368083","url":null,"abstract":"We investigate the use of genetic algorithms to evolve AI players for real-time strategy games. To overcome the knowledge acquisition bottleneck found in using traditional expert systems, scripts, or decision trees we evolve players through co-evolution. Our game players are implemented as resource allocation systems. Influence map trees are used to analyze the game-state and determine promising places to attack, defend, etc. These spatial objectives are chained to non-spatial objectives (train units, build buildings, gather resources) in a dependency graph. Players are encoded within the individuals of a genetic algorithm and co-evolved against each other, with results showing the production of strategies that are innovative, robust, and capable of defeating a suite of hand-coded opponents","PeriodicalId":365269,"journal":{"name":"2007 IEEE Symposium on Computational Intelligence and Games","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129965587","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：