arXivDaily arXiv每日学术速递 周一至周五更新
arXiv cs/0507044cs.LG
收藏

防御性通用学习与专家

Defensive Universal Learning with Experts

Jan Poland, Marcus Hutter

摘要:

本文展示了如何通过专家建议实现通用学习。为此,我们指定了一个专家算法,具有以下特点:(a)它只使用实际选择的行动(强盗设置)的反馈,(b)它可以应用于可数无限的专家类,(c)它处理的损失,可能会在适当的时间缓慢增长。我们证明了一个自适应对手的损失范围。由此,我们得到了一个主算法的“反应性”专家的问题,这意味着主的行动可能会影响对手的行为。我们的算法可以显着优于标准的专家算法在这样的问题。最后,我们将其与通用专家类相结合。由此产生的通用学习器执行-在某种意义上-几乎和任何可计算策略一样好,对于任何在线决策问题。我们还指定了(最坏情况下)收敛速度,这是非常慢的。

英文摘要:

This paper shows how universal learning can be achieved with expert advice. To this aim, we specify an experts algorithm with the following characteristics: (a) it uses only feedback from the actions actually chosen (bandit setup), (b) it can be applied with countably infinite expert classes, and (c) it copes with losses that may grow in time appropriately slowly. We prove loss bounds against an adaptive adversary. From this, we obtain a master algorithm for "reactive" experts problems, which means that the master's actions may influence the behavior of the adversary. Our algorithm can significantly outperform standard experts algorithms on such problems. Finally, we combine it with a universal expert class. The resulting universal learner performs -- in a certain sense -- almost as well as any computable strategy, for any online decision problem. We also specify the (worst-case) convergence speed, which is very slow.

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。

↑