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 2102.03802cs.LGmath.STstat.MLstat.TH
收藏

非线性度量学习的无维广义界

Dimension Free Generalization Bounds for Non Linear Metric Learning

Technion, Israel Institute of Technology(以色列理工学院)

Mark Kozdoba, Shie Mannor

摘要:

在这项工作中,我们研究的度量学习问题,其中的度量是由一个神经网络型嵌入的数据诱导的泛化保证。具体来说,我们提供了两个制度-稀疏制度,和一个非稀疏制度,我们的术语\emph{bounded amplification}统一的泛化界限。稀疏状态界限对应于参数的$\ell_1$型范数较小的情况。与分类中的情况类似,满足这样的界限的解可以通过问题的适当正则化来获得。另一方面,度量学习损失的非正则化SGD优化通常不会产生稀疏解。我们表明,尽管缺乏这种稀疏性,依靠不同的,新的属性的解决方案,它仍然是可能的,以提供无维泛化保证。因此,这些界限可以解释在非稀疏的真实实验情况下的泛化。我们说明了MNIST和20新闻组数据集上的研究现象。

英文摘要:

In this work we study generalization guarantees for the metric learning problem, where the metric is induced by a neural network type embedding of the data. Specifically, we provide uniform generalization bounds for two regimes -- the sparse regime, and a non-sparse regime which we term \emph{bounded amplification}. The sparse regime bounds correspond to situations where $\ell_1$-type norms of the parameters are small. Similarly to the situation in classification, solutions satisfying such bounds can be obtained by an appropriate regularization of the problem. On the other hand, unregularized SGD optimization of a metric learning loss typically does not produce sparse solutions. We show that despite this lack of sparsity, by relying on a different, new property of the solutions, it is still possible to provide dimension free generalization guarantees. Consequently, these bounds can explain generalization in non sparse real experimental situations. We illustrate the studied phenomena on the MNIST and 20newsgroups datasets.

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。

↑