Gerrit Anders, Christian Hinrichs, Florian Siefert, Pascal Behrmann, W. Reif, M. Sonnenschein
Multi-agent systems often consist of heterogeneous agents with different capabilities and objectives. While some agents might try to maximize their system's utility, others might be self-interested and thus only act for their own good. However, because of their limited capabilities and resources, it is often necessary that agents cooperate to be able to satisfy given tasks. To work together on such a task, the agents have to solve a task allocation problem, e.g., by teaming up in groups like coalitions or distributing the task among themselves on electronic markets. In this paper, we introduce two algorithms that allow agents to cooperatively solve a dynamic task allocation problem in uncertain environments. Based on these algorithms, we investigate the influence of inter-agent variation on the system's behavior. One of these algorithms explicitly exploits inter-agent variation to solve the task without communication between the agents, while the other builds upon a fixed overlay network in which agents exchange information. Throughout the paper, the frequency stabilization problem from the domain of decentralized power management serves as a running example to illustrate our algorithms and results.
{"title":"On the Influence of Inter-Agent Variation on Multi-Agent Algorithms Solving a Dynamic Task Allocation Problem under Uncertainty","authors":"Gerrit Anders, Christian Hinrichs, Florian Siefert, Pascal Behrmann, W. Reif, M. Sonnenschein","doi":"10.1109/SASO.2012.16","DOIUrl":"https://doi.org/10.1109/SASO.2012.16","url":null,"abstract":"Multi-agent systems often consist of heterogeneous agents with different capabilities and objectives. While some agents might try to maximize their system's utility, others might be self-interested and thus only act for their own good. However, because of their limited capabilities and resources, it is often necessary that agents cooperate to be able to satisfy given tasks. To work together on such a task, the agents have to solve a task allocation problem, e.g., by teaming up in groups like coalitions or distributing the task among themselves on electronic markets. In this paper, we introduce two algorithms that allow agents to cooperatively solve a dynamic task allocation problem in uncertain environments. Based on these algorithms, we investigate the influence of inter-agent variation on the system's behavior. One of these algorithms explicitly exploits inter-agent variation to solve the task without communication between the agents, while the other builds upon a fixed overlay network in which agents exchange information. Throughout the paper, the frequency stabilization problem from the domain of decentralized power management serves as a running example to illustrate our algorithms and results.","PeriodicalId":126067,"journal":{"name":"2012 IEEE Sixth International Conference on Self-Adaptive and Self-Organizing Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130154296","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}
The dynamics of collective attention emerging out of individual viewing experiences from public displays appear to be among the most demanding challenges in understanding the mechanisms of self-adaptation of public opinion. In this paper we approach a model of collective attention from observations of the attention of individuals estimated from their efforts expressing interest. Extending on SEEV, an established individual attention model from cognitive science, attention estimates from spontaneous passer-bys in front of public displays are used to describe a collective attention model at the scale of society. The model is validated via a large scale simulation experiment reflecting the demographics and the morphology of a whole city, together with population densities, mobility patterns and individual decision making on a 2048 node shared memory multiprocessor (SGI Altix Ultra Violet 1000, Repast HPC). Simulations how collective attention emerges from local spots of attention towards city scale opinion building and consensus finding.
{"title":"Collective Attention through Public Displays","authors":"A. Ferscha, K. Zia, Benedikt Gollan","doi":"10.1109/SASO.2012.35","DOIUrl":"https://doi.org/10.1109/SASO.2012.35","url":null,"abstract":"The dynamics of collective attention emerging out of individual viewing experiences from public displays appear to be among the most demanding challenges in understanding the mechanisms of self-adaptation of public opinion. In this paper we approach a model of collective attention from observations of the attention of individuals estimated from their efforts expressing interest. Extending on SEEV, an established individual attention model from cognitive science, attention estimates from spontaneous passer-bys in front of public displays are used to describe a collective attention model at the scale of society. The model is validated via a large scale simulation experiment reflecting the demographics and the morphology of a whole city, together with population densities, mobility patterns and individual decision making on a 2048 node shared memory multiprocessor (SGI Altix Ultra Violet 1000, Repast HPC). Simulations how collective attention emerges from local spots of attention towards city scale opinion building and consensus finding.","PeriodicalId":126067,"journal":{"name":"2012 IEEE Sixth International Conference on Self-Adaptive and Self-Organizing Systems","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133091620","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 a model to represent human activities in farming systems based on the situated action theory. The idea is to consider the environment as an intelligent entity that has the ability to decide locally, at given time and space, the actions to perform. Surrogates of real actors, called actuators, just execute the actions dictated by the agents embedded in the environment. The model is based on the affordance and stigmergy concepts as well as a multi-agents modelling approach.
{"title":"Modelling Situated Action Based on Affordances and Stigmergy","authors":"Zoubida Afoutni, F. Guerrin, R. Courdier","doi":"10.1109/SASO.2012.29","DOIUrl":"https://doi.org/10.1109/SASO.2012.29","url":null,"abstract":"This paper describes a model to represent human activities in farming systems based on the situated action theory. The idea is to consider the environment as an intelligent entity that has the ability to decide locally, at given time and space, the actions to perform. Surrogates of real actors, called actuators, just execute the actions dictated by the agents embedded in the environment. The model is based on the affordance and stigmergy concepts as well as a multi-agents modelling approach.","PeriodicalId":126067,"journal":{"name":"2012 IEEE Sixth International Conference on Self-Adaptive and Self-Organizing Systems","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116366595","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}
Ö. Gürcan, C. Bernon, K. Türker, J. Mano, P. Glize, Oğuz Dikenelli
Understanding functional synaptic connectivity of human central nervous system is one of the holy grails of the neuroscience. Due to the complexity of nervous system, it is common to reduce the problem to smaller networks such as motor unit pathways. In this sense, we designed and developed a simulation model that learns acting in the same way of human single motor units by using findings on human subjects. The developed model is based on self-organizing agents whose nominal and cooperative behaviors are based on the current knowledge on biological neural networks. The results show that the simulation model generates similar functionality with the observed data.
{"title":"Simulating Human Single Motor Units Using Self-Organizing Agents","authors":"Ö. Gürcan, C. Bernon, K. Türker, J. Mano, P. Glize, Oğuz Dikenelli","doi":"10.1109/SASO.2012.18","DOIUrl":"https://doi.org/10.1109/SASO.2012.18","url":null,"abstract":"Understanding functional synaptic connectivity of human central nervous system is one of the holy grails of the neuroscience. Due to the complexity of nervous system, it is common to reduce the problem to smaller networks such as motor unit pathways. In this sense, we designed and developed a simulation model that learns acting in the same way of human single motor units by using findings on human subjects. The developed model is based on self-organizing agents whose nominal and cooperative behaviors are based on the current knowledge on biological neural networks. The results show that the simulation model generates similar functionality with the observed data.","PeriodicalId":126067,"journal":{"name":"2012 IEEE Sixth International Conference on Self-Adaptive and Self-Organizing Systems","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115510423","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}
Fast and precise demand shaping is critical for the electrical power grid. With residential and small-business customers, a distributed approach to demand shaping is desirable for reasons of scalability and of privacy. The Color Power architecture [1] provides such an approach, but the controller previously used was badly limited. We now present an improved control algorithm, Color Power 2.0, based on stochastic constraint satisfaction, which provides major improvements in capability and performance over the prior algorithm. Analysis shows that its performance is within a small constant factor of optimal, and these results are confirmed empirically on simulated networks of 100 to 1 million devices.
{"title":"Fast Precise Distributed Control for Energy Demand Management","authors":"J. Beal, J. Berliner, Kevin Hunter","doi":"10.1109/SASO.2012.12","DOIUrl":"https://doi.org/10.1109/SASO.2012.12","url":null,"abstract":"Fast and precise demand shaping is critical for the electrical power grid. With residential and small-business customers, a distributed approach to demand shaping is desirable for reasons of scalability and of privacy. The Color Power architecture [1] provides such an approach, but the controller previously used was badly limited. We now present an improved control algorithm, Color Power 2.0, based on stochastic constraint satisfaction, which provides major improvements in capability and performance over the prior algorithm. Analysis shows that its performance is within a small constant factor of optimal, and these results are confirmed empirically on simulated networks of 100 to 1 million devices.","PeriodicalId":126067,"journal":{"name":"2012 IEEE Sixth International Conference on Self-Adaptive and Self-Organizing Systems","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129785399","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 conceive new self-organisation mechanisms to enhance the Gradient self-organisation pattern with anticipative adaptation abilities. We ensure that the problem of retrieving a target of interest in mobile environments is solved by proactively reacting to locally-available information about future events, namely, the knowledge about future obstacles (e.g., expected jams or road interruption in a traffic control scenario) is used to compute alternative and faster paths in an emergent way.
{"title":"Gradient-Based Self-Organisation Patterns of Anticipative Adaptation","authors":"Sara Montagna, Danilo Pianini, Mirko Viroli","doi":"10.1109/SASO.2012.25","DOIUrl":"https://doi.org/10.1109/SASO.2012.25","url":null,"abstract":"In this paper we conceive new self-organisation mechanisms to enhance the Gradient self-organisation pattern with anticipative adaptation abilities. We ensure that the problem of retrieving a target of interest in mobile environments is solved by proactively reacting to locally-available information about future events, namely, the knowledge about future obstacles (e.g., expected jams or road interruption in a traffic control scenario) is used to compute alternative and faster paths in an emergent way.","PeriodicalId":126067,"journal":{"name":"2012 IEEE Sixth International Conference on Self-Adaptive and Self-Organizing Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129918172","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}
Samhar Mahmoud, Daniel Villatoro, Jeroen Keppens, Michael Luck
The use of social norms has proven to be effective in the self-governance of decentralised systems in which there is no central authority. Axelrod's seminal model of norm establishment in populations of self-interested individuals provides some insight into the mechanisms needed to support this through the use of metanorms, but is not directly applicable to real world scenarios such as online peer-to-peer communities, for example. In particular, it does not reflect different topological arrangements of interactions. While some recent efforts have sought to address these limitations, they are also limited in not considering the point-to-point interactions between agents that arise in real systems, but only interactions that are visible to an entire neighbourhood. The objective of this paper is twofold: firstly to incorporate these realistic adaptations to the original model, and secondly, to provide agents with reputation based mechanisms that allow them to dynamically optimise the intensity of punishment ensuring norm establishment in exactly these limited observation conditions.
{"title":"Optimised Reputation-Based Adaptive Punishment for Limited Observability","authors":"Samhar Mahmoud, Daniel Villatoro, Jeroen Keppens, Michael Luck","doi":"10.1109/SASO.2012.24","DOIUrl":"https://doi.org/10.1109/SASO.2012.24","url":null,"abstract":"The use of social norms has proven to be effective in the self-governance of decentralised systems in which there is no central authority. Axelrod's seminal model of norm establishment in populations of self-interested individuals provides some insight into the mechanisms needed to support this through the use of metanorms, but is not directly applicable to real world scenarios such as online peer-to-peer communities, for example. In particular, it does not reflect different topological arrangements of interactions. While some recent efforts have sought to address these limitations, they are also limited in not considering the point-to-point interactions between agents that arise in real systems, but only interactions that are visible to an entire neighbourhood. The objective of this paper is twofold: firstly to incorporate these realistic adaptations to the original model, and secondly, to provide agents with reputation based mechanisms that allow them to dynamically optimise the intensity of punishment ensuring norm establishment in exactly these limited observation conditions.","PeriodicalId":126067,"journal":{"name":"2012 IEEE Sixth International Conference on Self-Adaptive and Self-Organizing Systems","volume":"179 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117226960","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}
Collective problem solving can lead to the development of new methods and algorithms that can potentially contribute to novel Artificial Intelligence applications and tools. Socially-inspired optimization algorithms are a class of algorithms that aim at conducting a search over a large solution space using mechanisms similar to how humans solve problems in a social context. Several such algorithms exist in the literature, including adaptations of classical ones, such as Genetic Algorithms. These models, however, do not take into account a fundamental concept in human social systems: the individual ability to adapt problem-solving strategies as a function of the social context. In this paper, we propose and investigate an extension inside a socially-inspired model of collective problem solving which allows one to model agents with such adaptability. This extension is based on the concept of humans as ``motivated tacticians'' and it dictates how agents are to adapt their search heuristics according to their respective social context. We show how this rule can speed up the system's convergence to good solutions and improve the search space exploration. The results contribute towards the design of socially inspired computational systems for collective problem-solving.
{"title":"Modeling Adaptative Social Behavior in Collective Problem Solving Algorithms","authors":"Diego Noble, L. Lamb, R. M. Araújo","doi":"10.1109/SASO.2012.20","DOIUrl":"https://doi.org/10.1109/SASO.2012.20","url":null,"abstract":"Collective problem solving can lead to the development of new methods and algorithms that can potentially contribute to novel Artificial Intelligence applications and tools. Socially-inspired optimization algorithms are a class of algorithms that aim at conducting a search over a large solution space using mechanisms similar to how humans solve problems in a social context. Several such algorithms exist in the literature, including adaptations of classical ones, such as Genetic Algorithms. These models, however, do not take into account a fundamental concept in human social systems: the individual ability to adapt problem-solving strategies as a function of the social context. In this paper, we propose and investigate an extension inside a socially-inspired model of collective problem solving which allows one to model agents with such adaptability. This extension is based on the concept of humans as ``motivated tacticians'' and it dictates how agents are to adapt their search heuristics according to their respective social context. We show how this rule can speed up the system's convergence to good solutions and improve the search space exploration. The results contribute towards the design of socially inspired computational systems for collective problem-solving.","PeriodicalId":126067,"journal":{"name":"2012 IEEE Sixth International Conference on Self-Adaptive and Self-Organizing Systems","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131215049","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}
Inter-related computations must sometimes maintain spatial relationships as they attempt to individually respond to changes in the environment. These changes may be related to application requirements (as in a sensor network) or resource requirements. Notably, maintenance of any spatial relationship is typically hard coded into application software using mechanism which roughly fit the mobile agent framework. Here we propose an alternative which allows application programmers to declaratively specify spatial relationships between related computations, which are then maintained by the run-time system. This paper argues for and presents key mechanisms involved in maintaining an important class of spatial relationships between computations.
{"title":"Maintaining Spatial Relationships in Uncertain Environments","authors":"Nadeem Jamali, Anil Keela","doi":"10.1109/SASO.2012.39","DOIUrl":"https://doi.org/10.1109/SASO.2012.39","url":null,"abstract":"Inter-related computations must sometimes maintain spatial relationships as they attempt to individually respond to changes in the environment. These changes may be related to application requirements (as in a sensor network) or resource requirements. Notably, maintenance of any spatial relationship is typically hard coded into application software using mechanism which roughly fit the mobile agent framework. Here we propose an alternative which allows application programmers to declaratively specify spatial relationships between related computations, which are then maintained by the run-time system. This paper argues for and presents key mechanisms involved in maintaining an important class of spatial relationships between computations.","PeriodicalId":126067,"journal":{"name":"2012 IEEE Sixth International Conference on Self-Adaptive and Self-Organizing Systems","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131585918","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}
A complex product is generally a system composed of numerous interdependent components, each one representing specific disciplines and developed using associated expertise. When analysing the problem from another point of view, we can see that for each design domain, a generally huge set of real already designed elements exists. Thus, when constructing a new element, it is interesting to use this already known and acquired knowledge. This knowledge does not only contain the discipline's information but also the engineers' experience. Considering this point of view, the design of complex products defines a new generic class of complex problems. In this paper, we address this class of problems using the Self-Adaptive Population Based Reasoning (SAPBR) generic approach. It is based on the Adaptive Multi-Agent System (AMAS) theory that takes advantage from cooperation to design robust and open multi-agent systems. In SAPBR, agents use cooperative self-tuning principles in order to estimate and discover new characteristic values for the design of new elements. The obtained system is compared to the Self-Organising Map (SOM) and the Multilayer Perceptron (MP) algorithms that address similar problems.
{"title":"Collective Self-Tuning for Complex Product Design","authors":"Elsy Kaddoum, J. Georgé","doi":"10.1109/SASO.2012.14","DOIUrl":"https://doi.org/10.1109/SASO.2012.14","url":null,"abstract":"A complex product is generally a system composed of numerous interdependent components, each one representing specific disciplines and developed using associated expertise. When analysing the problem from another point of view, we can see that for each design domain, a generally huge set of real already designed elements exists. Thus, when constructing a new element, it is interesting to use this already known and acquired knowledge. This knowledge does not only contain the discipline's information but also the engineers' experience. Considering this point of view, the design of complex products defines a new generic class of complex problems. In this paper, we address this class of problems using the Self-Adaptive Population Based Reasoning (SAPBR) generic approach. It is based on the Adaptive Multi-Agent System (AMAS) theory that takes advantage from cooperation to design robust and open multi-agent systems. In SAPBR, agents use cooperative self-tuning principles in order to estimate and discover new characteristic values for the design of new elements. The obtained system is compared to the Self-Organising Map (SOM) and the Multilayer Perceptron (MP) algorithms that address similar problems.","PeriodicalId":126067,"journal":{"name":"2012 IEEE Sixth International Conference on Self-Adaptive and Self-Organizing Systems","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128891783","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：