arXivDaily arXiv每日学术速递 周一至周五更新
arXiv cs/0508073cs.LGcs.AI
收藏

重复矩阵对策的通用学习

Universal Learning of Repeated Matrix Games

Jan Poland, Marcus Hutter

摘要:

我们研究和比较两个通用的学习算法,一个基于贝叶斯学习和预测专家的意见的学习动力学。这两种方法都有很强的渐近性能保证。当面临在重复的2x2矩阵游戏中找到好的长期策略的任务时,他们的表现完全不同。

英文摘要:

We study and compare the learning dynamics of two universal learning algorithms, one based on Bayesian learning and the other on prediction with expert advice. Both approaches have strong asymptotic performance guarantees. When confronted with the task of finding good long-term strategies in repeated 2x2 matrix games, they behave quite differently.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑