期刊文献+

基于Agent联盟的协作学习系统研究 被引量:1

Collaborative Learning System Based on Coalitions of Agents
下载PDF
导出
摘要 为了在协作学习系统中实现学习者Agent之间的有效合作,通过引入一种新的合作机制——同学关系网模型(School mate Relation Web Model),来构建学习系统中学习者Agent之间的同学联盟,并且基于学习者Agent之间的同学联盟来实现多个学习者Agent之间的协作学习。在每个同学联盟中,任意两个Agent之间都具有同学关系,并且联盟中的所有Agent相互协作,共同完成学习任务。另外,联盟中的学习者Agent之间的通信不是直接进行的,而是通过一个黑板来进行,这可以显著地提高Agent之间的通信效率。由于同学关系网模型可以避免Agent联盟形成的盲目性,并且可以提高学习者Agent之间的交互效率,从而使得我们基于Agent同学联盟的协作学习系统可以实现学习者Agent之间的有效合作,弥补了现有协作学习系统的不足。 In order to implement an efficient collaboration between different learners in a collaborative learning system, we proposed a new kind of collaborative mechanism -- Schoolmate Relation Web Model, by which we can construct coalitions of schoolmates among different learner Agents. Based on this kind of coalition of schoolmates we can implement the collaborative learning between different learner Agents. In each coalition of schoolmates, every Agent has a schoolmate relation with other Agents, they cooperate with each other and finish learning tasks together. And the agents in the coalition do not communicate directly with each other, but through a blackboard set up in the coalition, which can remarkably improve the efficiency of communication among Agents. Since Schoolmate Relation Web Model can avoid blindness in the forming process of coalition and improve the efficiency of negotiation among learner Agents,our collaborative learning system based on coalitions of schoolmates can realize an effective collaboration among learner agents and solve the problem of large communication and calculation cost which widely exists in current collaborative learning systems.
出处 《计算机科学》 CSCD 北大核心 2009年第6期125-128,共4页 Computer Science
基金 国家自然科学基金(60802042,60641010,60573063和60573064) 863计划项目(2007AA01Z325) 青岛市科技发展计划(07-2-3-3-jch) 青岛科技大学引进人才启动基金(200702583)资助
关键词 协作学习 多AGENT系统 远程教育 AGENT联盟 同学联盟 同学关系网 Collaborative learning, Multi-agent system, E-Learning, Coalition of agents, Coalition of schoolmates, Schoolmate relation Web
  • 相关文献

参考文献12

二级参考文献36

  • 1黄荣杯.CSCL的理论与方法[J].电化教育研究,1999,20(6):25-30. 被引量:101
  • 2叶斌,马忠贵,曾广平,尹怡欣,涂序彦.多Agent协商行为的效用分析[J].控制与决策,2004,19(12):1332-1336. 被引量:5
  • 3Smith R G. The contract net protocol: high-level communication and control in a distributed problem solver[J]. IEEE Trans. On Computers, 1980, C-29(12) :1104-1113.
  • 4Zlotkin G and Rosenschein J S. Coalition, cryptography, and stability: mechanisms for coalition formation in task oriented domains [C]. In Proceedings of AAAI94, Seattle, Washington.1994.
  • 5Shehory O and Kraus S. Methods for task allocation via agent coalition formation [J]. Artificial Intelligence 1998, 101 (1-2):165-200.
  • 6Shehory O and Kraus S. Coalition formation among autonomous agents: strategies and complexity[C]. In Proceedings of MAAMAW-93, Neuchatel, 1993.
  • 7Sandholm T W and Lesser V R. Coalition among computationally bounded agents [J]. Artificial Intelligence 1997,94: 99- 137.
  • 8Shehory O, Kraus S. Coalition formation among autonomous agents: Strategies and complexity [A]. Pmuller J. From Reaction to Cognition, Lecture Notes in Artificial Intelligence[C]. Berlin: Springer, 1993.57-72.
  • 9Shehory O, Kraus S. Task allocation via coalition formation among autonomous agents [Z]. The International Joint Conference on Artificial Intelligence, Montreal, 1995.
  • 10Pawlak Z. Why rough sets? [Z]. The Fifth IEEE International Conference on Fuzzy Systems,[s.l.], 1996.

共引文献181

同被引文献5

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部