arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2111.03044cs.LG
收藏

Coreset学习的统一方法

A Unified Approach to Coreset Learning

University of Haifa(海法大学)

Alaa Maalouf, Gilad Eini, Ben Mussay, Dan Feldman, Margarita Osadchy

摘要:

给定数据集和损失函数的核心集通常是一个小的加权集,它近似于给定查询集中每个查询的损失。核心集在许多应用中非常有用。然而,核心集的构造是以问题相关的方式完成的,并且可能需要数年的时间来设计和证明特定查询族的核心集的正确性。这可能会限制coresets在实际应用中的使用。此外,小核心集可证明不存在许多问题。为了解决这些限制,我们提出了一个通用的,基于学习的算法建设coresets。我们的方法提供了一个新的定义coreset,这是一个自然放松的标准定义,旨在近似的\emph{average}损失的原始数据的查询。这允许我们使用一个学习范例来计算一个给定的输入集相对于一个给定的损失函数使用一个训练集的查询的小coreset。我们得到正式的保证所提出的方法。对深度网络和经典机器学习问题的实验评估表明,我们学习的核心集产生了与现有算法相当甚至更好的结果,具有最坏情况的理论保证(在实践中可能过于悲观)。此外,我们应用于深度网络修剪的方法为完整的深度网络提供了第一个核心集,即,一次压缩所有网络,而不是逐层或类似的分治方法。

英文摘要:

Coreset of a given dataset and loss function is usually a small weighed set that approximates this loss for every query from a given set of queries. Coresets have shown to be very useful in many applications. However, coresets construction is done in a problem dependent manner and it could take years to design and prove the correctness of a coreset for a specific family of queries. This could limit coresets use in practical applications. Moreover, small coresets provably do not exist for many problems. To address these limitations, we propose a generic, learning-based algorithm for construction of coresets. Our approach offers a new definition of coreset, which is a natural relaxation of the standard definition and aims at approximating the \emph{average} loss of the original data over the queries. This allows us to use a learning paradigm to compute a small coreset of a given set of inputs with respect to a given loss function using a training set of queries. We derive formal guarantees for the proposed approach. Experimental evaluation on deep networks and classic machine learning problems show that our learned coresets yield comparable or even better results than the existing algorithms with worst-case theoretical guarantees (that may be too pessimistic in practice). Furthermore, our approach applied to deep network pruning provides the first coreset for a full deep network, i.e., compresses all the network at once, and not layer by layer or similar divide-and-conquer methods.

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。

↑