arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.02605cs.LGstat.ML
收藏

迈向大规模核模型

Toward Large Kernel Models

UC San Diego(加州大学圣地亚哥分校)

Amirhesam Abedsoltan, Mikhail Belkin, Parthe Pandit

AI总结:

本文提出EigenPro 3.0算法,基于投影对偶预条件SGD,解耦核机器模型与数据大小,实现大规模通用核模型的训练,突破现有核方法在模型和数据规模上的限制。

摘要:

近期研究表明,在小数据集上,核机器通常可以表现得与深度神经网络(DNN)相当或更好。核机器的吸引力还因其在某些情况下与宽神经网络的等价性的发现而进一步增强。然而,深度神经网络的一个关键特征是能够独立地扩展模型大小和训练数据大小,而在传统核机器中,模型大小与数据大小绑定。由于这种耦合,将核机器扩展到大数据在计算上一直具有挑战性。在本文中,我们为构建大规模通用核模型提供了一条前进之路,该模型是核机器的一种推广,解耦了模型与数据,从而允许在大数据集上进行训练。具体而言,我们引入了EigenPro 3.0,一种基于投影对偶预条件随机梯度下降的算法,并展示了其在模型和数据规模上的扩展能力,这是现有核方法无法实现的。

英文摘要:

Recent studies indicate that kernel machines can often perform similarly or better than deep neural networks (DNNs) on small datasets. The interest in kernel machines has been additionally bolstered by the discovery of their equivalence to wide neural networks in certain regimes. However, a key feature of DNNs is their ability to scale the model size and training data size independently, whereas in traditional kernel machines model size is tied to data size. Because of this coupling, scaling kernel machines to large data has been computationally challenging. In this paper, we provide a way forward for constructing large-scale general kernel models, which are a generalization of kernel machines that decouples the model and data, allowing training on large datasets. Specifically, we introduce EigenPro 3.0, an algorithm based on projected dual preconditioned SGD and show scaling to model and data sizes which have not been possible with existing kernel 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。

↑