HH Bui, S Venkatesh and DH Kieronska, Learning Other Agents' Preferences in Multi-Agent Negotiation Using the Bayesian Classifier. International Journal of Cooperative Information Systems (IJCIS) , Vol. 8, No. 4, 1999, pp. 275--294.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....on agent platform) and adaptation of an agent s functionality. Research on adaptation of knowledge and facts of an agent is usually based on (machine) learning, e.g. 3] Example applications include personification: an agent maintains and adapts a profile of its (human) clients, e.g. [4], 5] and [6] co ordination in multi agent systems, e.g. 7] and [8] and situated learning for agents, e.g. 9] Research on adaptation of the interface of an agent is usually concerned with adapting the agent s interface to the (current) agent platform, e.g. see [10] 11] Research on ....

Bui, H.H., Kieronska, D., and Venkatesh, S.: Learning Other Agents' Preferences in Multiagent Negotiation. In: Proceedings of the National Conference on Artificial Intelligence (AAAI-96) (1996) 114--119

....and learning agents are often employed to maintain profiles of (human) agents. For example, a consumer agent maintains a profile of the personal assistant agent, and adapts this profile on the basis of interaction with the personal assistant agent (e.g. as also encountered in negotiation settings [10,12]) Adaptation may entail mobility of an agent: an agent may migrate to a specific place to access locally available resources. Agents, as studied within AI, are (to some extent) intelligent, and interact to solve complex problems. The overall behaviour of multi agent systems is an important ....

H. H. Bui, D. Kieronska, and S. Venkatesh. Learning other agents' preferences in multiagent negotiation. In Proceedings of the National Conference on Artificial Intelligence (AAAI-96), pages 114--119, 1996.

.... We provide a learning agent architecture that integrates a learning component into a reactive agent architecture in which the agents negotiate by refining a joint intention gradually until a common consensus is reached [4] We also extend the simple instance averaging learning mechanism in [3] to a more general Bayesian classification mechanism based on which the learning component is implemented. Learning gives the agent the ability to draw predictions in uncertain situations, thus reducing the pressure for communication and improving the overall efficiency of the negotiation process. ....

....Online learning tasks are best suited to some form of incremental learning to avoid storing the ever growing, full accumulated sample set and keep updating time to a minimum. We have experimented with one form of incremental learning, the instance averaging mechanism, in our previous work [3]. However, the mechanism has several short comings, among which is the inability to learn complex patterns that might exist in other agents preference functions. To over come this, the (naive) Bayesian classifiers are used here as the underlying learning mechanism. In the rest of this section, we ....

Hung H. Bui, D. Kieronska, and S. Venkatesh. Learning other agents' preferences in multiagent negotiation. In Proceedings of the National Conference on Artificial Intelligence (AAAI-96), pages 114--119, August 1996.

.... We provide a learning agent architecture that integrates a learning component into a reactive agent architecture in which the agents negotiate by refining a joint intention gradually until a common consensus is reached [4] We also extend the simple instance averaging learning mechanism in [3] to a more general Bayesian classification mechanism based on which the learning component is implemented. Learning gives the agent the ability to draw predictions in uncertain situations, thus reducing the pressure for communication and improving the overall efficiency of the negotiation process. ....

....Online learning tasks are best suited to some form of incremental learning to avoid storing the ever growing, full accumulated sample set and keep updating time to a minimum. We have experimented with one form of incremental learning, the instance averaging mechanism, in our previous work [3]. However, the mechanism has several short comings, among which is the inability to learn complex patterns that might exist in other agents preference functions. To over come this, the (naive) Bayesian classifiers are used here as the underlying learning mechanism. In the rest of this section, we ....

Hung H. Bui, D. Kieronska, and S. Venkatesh. Learning other agents' preferences in multiagent negotiation. In Proceedings of the National Conference on Artificial Intelligence (AAAI-96), pages 114--119, August 1996.

....optimal. Our work in integrating learning into the framework of team coordination is related to the recent body of work in multi agent learning [15, 19] In team setting, groups of learning agents have been shown to achieve better performance than groups of non learning agents in various domains [2, 5, 11]. Our work here can serve as a theoretical model of such systems of learning agents. The rest of the paper is organised as follows. In the next section, we present the framework for coordination in team problems with incomplete information. Next, we enrich the framework with a model for learning ....

H. H. Bui, D. Kieronska, and S. Venkatesh. Learning other agents' preferences in multiagent negotiation. In Proceedings of the National Conference on Artificial Intelligence (AAAI-96), pages 114--119, August 1996.

No context found.

HH Bui, S Venkatesh and DH Kieronska, Learning Other Agents' Preferences in Multi-Agent Negotiation Using the Bayesian Classifier. International Journal of Cooperative Information Systems (IJCIS) , Vol. 8, No. 4, 1999, pp. 275--294.

No context found.

H. Bui, D. Kieronska, and S. Venkatesh. Learning other agents' preferences in multiagent negotiation. In Proceedings of the Thirteenth National Conference on Artificial Intelligence, pages 114--119, Menlo Park, CA, 1996. AAAI Press.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC