Pub Date : 2001-10-30DOI: 10.1109/ICCIMA.2001.970497
M. Venkatesan, H. Selvaraj, R. Bignall
Summary form only given. Functional decomposition is a process of representing a complex function as a function of functions with fewer variables. Earlier partition based functional decomposition tools represent the functions using r-partition. The r-partition representation is an abstract representation of the function and their memory requirements are super-exponential. An improved functional representation called ir-partition is proposed. The ir-partition representation is a complete representation of the function and requires less memory to store the functions. The main idea behind the ir-partition representation is to incorporate the values of the minterms corresponding to the variables (cubes). Hence, repeated access of the truth table is not necessary to read the value of the minterms. The computational time to calculate the ir-partition operations are three times greater than the computational time and memory requirement to calculate r-partition. However, the memory requirements for representing the function using ir-partition is half the memory requirement using the r-partition representation (abstract representation). Their partition representation also allows us to perform certain Partition Calculus operations implicitly. The representation has been implemented and tested with the MCNC benchmarks.
{"title":"An improved representation of functions for partition based functional decomposition","authors":"M. Venkatesan, H. Selvaraj, R. Bignall","doi":"10.1109/ICCIMA.2001.970497","DOIUrl":"https://doi.org/10.1109/ICCIMA.2001.970497","url":null,"abstract":"Summary form only given. Functional decomposition is a process of representing a complex function as a function of functions with fewer variables. Earlier partition based functional decomposition tools represent the functions using r-partition. The r-partition representation is an abstract representation of the function and their memory requirements are super-exponential. An improved functional representation called ir-partition is proposed. The ir-partition representation is a complete representation of the function and requires less memory to store the functions. The main idea behind the ir-partition representation is to incorporate the values of the minterms corresponding to the variables (cubes). Hence, repeated access of the truth table is not necessary to read the value of the minterms. The computational time to calculate the ir-partition operations are three times greater than the computational time and memory requirement to calculate r-partition. However, the memory requirements for representing the function using ir-partition is half the memory requirement using the r-partition representation (abstract representation). Their partition representation also allows us to perform certain Partition Calculus operations implicitly. The representation has been implemented and tested with the MCNC benchmarks.","PeriodicalId":232504,"journal":{"name":"Proceedings Fourth International Conference on Computational Intelligence and Multimedia Applications. ICCIMA 2001","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129331036","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 : 2001-10-30DOI: 10.1109/ICCIMA.2001.970466
M. Kinoshita, H. Yokoi, Y. Kakazu, Michiko Watanabe, T. Kawakami
It is very difficult to estimate behaviors of multiple autonomous robots or mutual interactions of them in real time. Therefore, we propose a quantitative observation approach of multiple robots behaviors. This approach introduces thermodynamic macroscopic state values to the multi-robot systems. The advantage of this approach is that it enables to observe the behaviors of autonomous robots in real world and can be mapped to characteristic values in another conceptual state space. Thermodynamic macroscopic state values, such as temperature, pressure and entropy, are defined in mobile robots systems. In our definition, each mobile robot is supposed to have a particle in thermodynamic systems. The experiment shows that the states of robots system can be classified by thermodynamic macroscopic state value. This verifies that the macroscopic quantitative observation is efficient and applicable to control multi-robot systems.
{"title":"Macroscopic quantitative observation of multi-robot behavior","authors":"M. Kinoshita, H. Yokoi, Y. Kakazu, Michiko Watanabe, T. Kawakami","doi":"10.1109/ICCIMA.2001.970466","DOIUrl":"https://doi.org/10.1109/ICCIMA.2001.970466","url":null,"abstract":"It is very difficult to estimate behaviors of multiple autonomous robots or mutual interactions of them in real time. Therefore, we propose a quantitative observation approach of multiple robots behaviors. This approach introduces thermodynamic macroscopic state values to the multi-robot systems. The advantage of this approach is that it enables to observe the behaviors of autonomous robots in real world and can be mapped to characteristic values in another conceptual state space. Thermodynamic macroscopic state values, such as temperature, pressure and entropy, are defined in mobile robots systems. In our definition, each mobile robot is supposed to have a particle in thermodynamic systems. The experiment shows that the states of robots system can be classified by thermodynamic macroscopic state value. This verifies that the macroscopic quantitative observation is efficient and applicable to control multi-robot systems.","PeriodicalId":232504,"journal":{"name":"Proceedings Fourth International Conference on Computational Intelligence and Multimedia Applications. ICCIMA 2001","volume":"116 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128082292","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 : 2001-10-30DOI: 10.1109/ICCIMA.2001.970465
X. Yao, Yong Liu
The paper studies the application of evolutionary artificial neural networks to chlorophyll-a prediction in Lake Kasumigaura (in Japan). Unlike previous applications of artificial neural networks in this field, the architecture of the artificial neural network is evolved automatically rather than designed manually. The evolutionary system is able to find a near optimal architecture of the artificial neural network for the prediction task. Our experimental results have shown that evolved artificial neural networks are very compact and generalise well. The evolutionary system is able to explore a large space of possible artificial neural networks and discover novel artificial neural networks for solving a problem.
{"title":"Evolving neural networks for chlorophyll-a prediction","authors":"X. Yao, Yong Liu","doi":"10.1109/ICCIMA.2001.970465","DOIUrl":"https://doi.org/10.1109/ICCIMA.2001.970465","url":null,"abstract":"The paper studies the application of evolutionary artificial neural networks to chlorophyll-a prediction in Lake Kasumigaura (in Japan). Unlike previous applications of artificial neural networks in this field, the architecture of the artificial neural network is evolved automatically rather than designed manually. The evolutionary system is able to find a near optimal architecture of the artificial neural network for the prediction task. Our experimental results have shown that evolved artificial neural networks are very compact and generalise well. The evolutionary system is able to explore a large space of possible artificial neural networks and discover novel artificial neural networks for solving a problem.","PeriodicalId":232504,"journal":{"name":"Proceedings Fourth International Conference on Computational Intelligence and Multimedia Applications. ICCIMA 2001","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129185342","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 : 2001-10-30DOI: 10.1109/ICCIMA.2001.970493
E. Angelotti, E. Scalabrin, B. C. Ávila
This work is part of the Multicheck Project that defines architecture of cognitive and independents agents for the automatic treatment of handwritten Brazilian bank checks. The concept of autonomous agents allows us to organize the application knowledge and brings several own benefits to the approach. The choice of this approach is supported in a triple hypothesis. First, the nature of the problem in question allows decomposition in well-defined tasks, and each of them can be encapsulated in an independent agent. Second, the natural capability of interaction of the agents makes the check treatment process more robust, solving situations apparently difficult. Third, the natural parallelism between the agents can contribute to implement an application with high performance.
{"title":"PANDORA: a multi-agent system using paraconsistent logic","authors":"E. Angelotti, E. Scalabrin, B. C. Ávila","doi":"10.1109/ICCIMA.2001.970493","DOIUrl":"https://doi.org/10.1109/ICCIMA.2001.970493","url":null,"abstract":"This work is part of the Multicheck Project that defines architecture of cognitive and independents agents for the automatic treatment of handwritten Brazilian bank checks. The concept of autonomous agents allows us to organize the application knowledge and brings several own benefits to the approach. The choice of this approach is supported in a triple hypothesis. First, the nature of the problem in question allows decomposition in well-defined tasks, and each of them can be encapsulated in an independent agent. Second, the natural capability of interaction of the agents makes the check treatment process more robust, solving situations apparently difficult. Third, the natural parallelism between the agents can contribute to implement an application with high performance.","PeriodicalId":232504,"journal":{"name":"Proceedings Fourth International Conference on Computational Intelligence and Multimedia Applications. ICCIMA 2001","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117212265","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 : 2001-10-30DOI: 10.1109/ICCIMA.2001.970499
L. Giraffa, M. Mora, A. Zamberlam
This paper shows some aspects of our work using a Multi-agent System applied to build interactive Intelligent Tutoring Systems (ITS) with student model based on mental states (BDI agents). We present some aspects about BDI agents' implementation, and the tool (E-BDI editor) under construction. We do believe that building BDI agents is not simple. This declarative paradigm and its implementation needs to be guided by a visual tool. This editor was designed to aid research to organise the basic set of mental states needed to model the cognitive agents.
{"title":"Working on student models (really) based on mental states","authors":"L. Giraffa, M. Mora, A. Zamberlam","doi":"10.1109/ICCIMA.2001.970499","DOIUrl":"https://doi.org/10.1109/ICCIMA.2001.970499","url":null,"abstract":"This paper shows some aspects of our work using a Multi-agent System applied to build interactive Intelligent Tutoring Systems (ITS) with student model based on mental states (BDI agents). We present some aspects about BDI agents' implementation, and the tool (E-BDI editor) under construction. We do believe that building BDI agents is not simple. This declarative paradigm and its implementation needs to be guided by a visual tool. This editor was designed to aid research to organise the basic set of mental states needed to model the cognitive agents.","PeriodicalId":232504,"journal":{"name":"Proceedings Fourth International Conference on Computational Intelligence and Multimedia Applications. ICCIMA 2001","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129746764","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 : 2001-10-30DOI: 10.1109/ICCIMA.2001.970467
T. Yamashita, H. Kawamura, M. Yamamoto, A. Ohuchi
The authors introduce a mutual choice mechanism into the norms game instead of direct penal regulation and reformulate the norms game with mutual choice. In order to examine how the proportion of metanorm players to all players influences the establishment of norm in the metanorms game with mutual choice, we perform simulation with several different proportions of metanorm players. We exclude the game theoretic assumption of uniformity as the metanorm player and observe the frequency of the establishment of the norm. As a result, we confirm the robustness of mutual choice against insufficiency of metanorm players.
{"title":"Effects of proportion of metanorm players on establishment of norm","authors":"T. Yamashita, H. Kawamura, M. Yamamoto, A. Ohuchi","doi":"10.1109/ICCIMA.2001.970467","DOIUrl":"https://doi.org/10.1109/ICCIMA.2001.970467","url":null,"abstract":"The authors introduce a mutual choice mechanism into the norms game instead of direct penal regulation and reformulate the norms game with mutual choice. In order to examine how the proportion of metanorm players to all players influences the establishment of norm in the metanorms game with mutual choice, we perform simulation with several different proportions of metanorm players. We exclude the game theoretic assumption of uniformity as the metanorm player and observe the frequency of the establishment of the norm. As a result, we confirm the robustness of mutual choice against insufficiency of metanorm players.","PeriodicalId":232504,"journal":{"name":"Proceedings Fourth International Conference on Computational Intelligence and Multimedia Applications. ICCIMA 2001","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129265038","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 : 2001-10-30DOI: 10.1109/ICCIMA.2001.970442
T. Iba, Y. Takabe, Y. Chubachi, Y. Takefuji
The authors propose a "Boxed Economy Simulation Platform", which is a sharable basis for agent-based economic simulations. By providing the basic design of the social model, which we call "Boxed Economy Foundation Model", it enables collaborative research more efficiently. Sharing and cumulating the model components can be promoted by domain-specific design at the level of social model rather than the level of abstract general purpose model. It will be able to contribute to remove factors that have been making it difficult for social scientists to participate in and conduct agent-based research.
{"title":"Boxed Economy Simulation Platform and foundation model","authors":"T. Iba, Y. Takabe, Y. Chubachi, Y. Takefuji","doi":"10.1109/ICCIMA.2001.970442","DOIUrl":"https://doi.org/10.1109/ICCIMA.2001.970442","url":null,"abstract":"The authors propose a \"Boxed Economy Simulation Platform\", which is a sharable basis for agent-based economic simulations. By providing the basic design of the social model, which we call \"Boxed Economy Foundation Model\", it enables collaborative research more efficiently. Sharing and cumulating the model components can be promoted by domain-specific design at the level of social model rather than the level of abstract general purpose model. It will be able to contribute to remove factors that have been making it difficult for social scientists to participate in and conduct agent-based research.","PeriodicalId":232504,"journal":{"name":"Proceedings Fourth International Conference on Computational Intelligence and Multimedia Applications. ICCIMA 2001","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130147014","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 : 2001-10-30DOI: 10.1109/ICCIMA.2001.970456
T. Yamaguchi, Ryo Marukawa
Presents a new framework of multi-agent reinforcement learning to acquire cooperative behaviors by generating and coordinating each learning goal interactively among agents. One of the main goals of artificial intelligence is to realize an intelligent agent that behaves autonomously by its sense of values. Reinforcement learning (RL) is the major learning mechanism for the agent to adapt itself to various situations of an unknown environment flexibly. However, in a multi-agent system environment that has mutual dependency among agents, it is difficult for a human to set up suitable learning goals for each agent, and, in addition, the existing framework of RL that aims for egoistic optimality of each agent is inadequate. Therefore, an active and interactive learning mechanism is required to generate and coordinate each learning goal among the agents. To realize this, first we propose to treat each learning goal as a reinforcement signal (RS) that can be communicated among the agents. Second, we introduce motivation rules to integrate the RSs communicated among the agents into a reward value for RL of an agent. Then we define cooperative rewards as learning goals with mutual dependency. Learning experiments for two agents with various motivation rules are performed. The experimental results show that several combinations of motivation rules converge to cooperative behaviors.
{"title":"Interactive multiagent reinforcement learning with motivation rules","authors":"T. Yamaguchi, Ryo Marukawa","doi":"10.1109/ICCIMA.2001.970456","DOIUrl":"https://doi.org/10.1109/ICCIMA.2001.970456","url":null,"abstract":"Presents a new framework of multi-agent reinforcement learning to acquire cooperative behaviors by generating and coordinating each learning goal interactively among agents. One of the main goals of artificial intelligence is to realize an intelligent agent that behaves autonomously by its sense of values. Reinforcement learning (RL) is the major learning mechanism for the agent to adapt itself to various situations of an unknown environment flexibly. However, in a multi-agent system environment that has mutual dependency among agents, it is difficult for a human to set up suitable learning goals for each agent, and, in addition, the existing framework of RL that aims for egoistic optimality of each agent is inadequate. Therefore, an active and interactive learning mechanism is required to generate and coordinate each learning goal among the agents. To realize this, first we propose to treat each learning goal as a reinforcement signal (RS) that can be communicated among the agents. Second, we introduce motivation rules to integrate the RSs communicated among the agents into a reward value for RL of an agent. Then we define cooperative rewards as learning goals with mutual dependency. Learning experiments for two agents with various motivation rules are performed. The experimental results show that several combinations of motivation rules converge to cooperative behaviors.","PeriodicalId":232504,"journal":{"name":"Proceedings Fourth International Conference on Computational Intelligence and Multimedia Applications. ICCIMA 2001","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122348762","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 : 2001-10-30DOI: 10.1109/ICCIMA.2001.970455
K. Miyazaki, S. Kobayashi
Reinforcement learning is a kind of machine learning. It aims to adapt an agent to an unknown environment according to rewards. Traditionally, from a theoretical point of view, many reinforcement learning systems assume that the environment has Markovian properties. However, it is important to treat non-Markovian environments in multi-agent reinforcement learning systems. The authors use Profit Sharing (PS) as a reinforcement learning system and discuss the rationality of PS in multi-agent environments. In particular, we classify non-Markovian environments and discuss how to share a reward among reinforcement learning agents. Through a crane control problem, we confirm the effectiveness of PS in multi-agent environments.
{"title":"On the rationality of profit sharing in multi-agent reinforcement learning","authors":"K. Miyazaki, S. Kobayashi","doi":"10.1109/ICCIMA.2001.970455","DOIUrl":"https://doi.org/10.1109/ICCIMA.2001.970455","url":null,"abstract":"Reinforcement learning is a kind of machine learning. It aims to adapt an agent to an unknown environment according to rewards. Traditionally, from a theoretical point of view, many reinforcement learning systems assume that the environment has Markovian properties. However, it is important to treat non-Markovian environments in multi-agent reinforcement learning systems. The authors use Profit Sharing (PS) as a reinforcement learning system and discuss the rationality of PS in multi-agent environments. In particular, we classify non-Markovian environments and discuss how to share a reward among reinforcement learning agents. Through a crane control problem, we confirm the effectiveness of PS in multi-agent environments.","PeriodicalId":232504,"journal":{"name":"Proceedings Fourth International Conference on Computational Intelligence and Multimedia Applications. ICCIMA 2001","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124240286","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 : 2001-10-30DOI: 10.1109/ICCIMA.2001.970486
J.M. Garcia-Ortegon, J. Torres-Jiménez
We present a set of data visualization tools that could lead to a better understanding of the conditions that make of a SAT instance a hard one. The visualization techniques included in this work are used to make evident the relationships between the SAT-variables in function of their distribution, signs and combinations of signs in the clauses. Using this information, the user may identify patterns associated with the hardness of a SAT instance which gives a more flexible measurement of the instance's hardness than just the relationship between the number of clauses (M) and the number of variables (N). It could be even possible to use the developed data visualization tools to determine if a specific solution method is best suitable for a specific SAT instance.
{"title":"Data visualization tools for 3SAT instances","authors":"J.M. Garcia-Ortegon, J. Torres-Jiménez","doi":"10.1109/ICCIMA.2001.970486","DOIUrl":"https://doi.org/10.1109/ICCIMA.2001.970486","url":null,"abstract":"We present a set of data visualization tools that could lead to a better understanding of the conditions that make of a SAT instance a hard one. The visualization techniques included in this work are used to make evident the relationships between the SAT-variables in function of their distribution, signs and combinations of signs in the clauses. Using this information, the user may identify patterns associated with the hardness of a SAT instance which gives a more flexible measurement of the instance's hardness than just the relationship between the number of clauses (M) and the number of variables (N). It could be even possible to use the developed data visualization tools to determine if a specific solution method is best suitable for a specific SAT instance.","PeriodicalId":232504,"journal":{"name":"Proceedings Fourth International Conference on Computational Intelligence and Multimedia Applications. ICCIMA 2001","volume":"234 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114257825","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：