arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1011.3168stat.MLcs.GTcs.LG
收藏

在线学习:超越遗憾

Online Learning: Beyond Regret

University of Pennsylvania(宾夕法尼亚大学);TTIC(丰田技术研究所(芝加哥));University of Texas at Austin(德克萨斯大学奥斯汀分校)

Alexander Rakhlin, Karthik Sridharan, Ambuj Tewari

摘要:

我们研究在线学习的广泛的一类问题,扩展(Rakhlin,Sridharan,Tewari,2010)的结果,以一般的概念,远远超出外部遗憾的性能测量。我们的框架同时捕捉这些著名的概念,如内部和一般的Phi遗憾,学习与非添加剂的全球成本函数,布莱克威尔的可接近性,校准的预测,自适应遗憾,等等。我们表明,在所有这些情况下的可学习性是由于控制相同的三个量:一个鞅收敛项,一个术语描述的能力,表现良好,如果未来是已知的,和序列的广义拉德马赫复杂度,研究(Rakhlin,Sridharan,Tewari,2010)。由于我们直接研究问题的复杂性,而不是专注于有效的算法,我们能够改进和扩展许多已知的结果,这些结果以前已经通过一个算法的建设。

英文摘要:

We study online learnability of a wide class of problems, extending the results of (Rakhlin, Sridharan, Tewari, 2010) to general notions of performance measure well beyond external regret. Our framework simultaneously captures such well-known notions as internal and general Phi-regret, learning with non-additive global cost functions, Blackwell's approachability, calibration of forecasters, adaptive regret, and more. We show that learnability in all these situations is due to control of the same three quantities: a martingale convergence term, a term describing the ability to perform well if future is known, and a generalization of sequential Rademacher complexity, studied in (Rakhlin, Sridharan, Tewari, 2010). Since we directly study complexity of the problem instead of focusing on efficient algorithms, we are able to improve and extend many known results which have been previously derived via an algorithmic construction.

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。

↑