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 1108.3154cs.LGstat.ML
收藏

在线学习的稳定性条件

Stability Conditions for Online Learnability

Carnegie Mellon University(卡内基梅隆大学)

Stephane Ross, J. Andrew Bagnell

摘要:

稳定性是量化学习算法的输出对训练数据集中的小变化(例如单个训练样本的删除或替换)的敏感性的一般概念。这种条件最近已被证明是更强大的特点,在一般的学习设置下i.i.d.一致收敛对于可学习性不是必需的,但是稳定性对于可学习性是充分和必要的。我们在这里表明,类似的稳定性条件也足以在线学习,即是否存在一个学习算法,使任何序列的例子(可能选择对抗)产生一个序列的假设,没有遗憾的限制,在事后的最佳假设。我们引入在线稳定性,稳定性条件有关的均匀留一稳定性在批量设置,这是足够的在线学习。特别是,我们表明,流行类的在线学习者,即算法属于以下类别的(正规化)领导者,镜像下降,基于梯度的方法和随机算法,如加权多数和对冲,保证没有遗憾,如果他们有这样的在线稳定性。我们提供的例子表明,存在这样的稳定性条件的算法实际上可能是必要的在线学习。对于更严格的二进制分类设置,我们建立这样的稳定性条件实际上是充分和必要的。我们还表明,在一般的学习设置,即那些与次指数覆盖的概念,没有遗憾的在线算法,有这样的稳定性条件存在的一大类在线学习的问题。

英文摘要:

Stability is a general notion that quantifies the sensitivity of a learning algorithm's output to small change in the training dataset (e.g. deletion or replacement of a single training sample). Such conditions have recently been shown to be more powerful to characterize learnability in the general learning setting under i.i.d. samples where uniform convergence is not necessary for learnability, but where stability is both sufficient and necessary for learnability. We here show that similar stability conditions are also sufficient for online learnability, i.e. whether there exists a learning algorithm such that under any sequence of examples (potentially chosen adversarially) produces a sequence of hypotheses that has no regret in the limit with respect to the best hypothesis in hindsight. We introduce online stability, a stability condition related to uniform-leave-one-out stability in the batch setting, that is sufficient for online learnability. In particular we show that popular classes of online learners, namely algorithms that fall in the category of Follow-the-(Regularized)-Leader, Mirror Descent, gradient-based methods and randomized algorithms like Weighted Majority and Hedge, are guaranteed to have no regret if they have such online stability property. We provide examples that suggest the existence of an algorithm with such stability condition might in fact be necessary for online learnability. For the more restricted binary classification setting, we establish that such stability condition is in fact both sufficient and necessary. We also show that for a large class of online learnable problems in the general learning setting, namely those with a notion of sub-exponential covering, no-regret online algorithms that have such stability condition exists.

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。

↑