arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2111.09459math.PRcs.LGstat.ML
收藏

图上梯度流:存在性、收敛性、连续方程

Gradient flows on graphons: existence, convergence, continuity equations

Sewoong Oh Paul G. Allen School of Computer Science \& Engineering University of Washington Seattle WA 98195, USA Email;Soumik Pal Department of Mathematics University of Washington Seattle WA 98195, USA Email;Raghav Somani Paul G. Allen School of Computer Science \& Engineering University of Washington Seattle WA 98195, USA Email;Raghavendra Tripathi Department of Mathematics University of Washington Seattle WA 98195, USA Email

Sewoong Oh, Soumik Pal, Raghav Somani, Raghavendra Tripathi

AI总结:

本文研究了图上梯度流的连续极限,证明了欧几里得梯度流在边权重上的收敛性,并探讨了图上空间上的最大坡度曲线。

摘要:

在概率测度上的Wasserstein梯度流已在各种优化问题中找到了广泛的应用。它们通常作为可交换粒子系统在某种均场相互作用中演化的连续极限。然而,在许多问题中,如多层神经网络,所谓的粒子是大型图上的边权重,其节点是可交换的。此类大型图在尺寸趋于无穷大时已知会收敛到称为图上(graphons)的连续极限。我们证明了欧几里得梯度流在合适函数的边权重上收敛到一个新的连续极限,该极限由图上空间上的曲线给出,可以适当描述为梯度流或更技术性地描述为最大坡度曲线。几种自然的图上函数,如同构函数和标量熵,都被我们的设定所涵盖,且例子已详细工作。

英文摘要:

Wasserstein gradient flows on probability measures have found a host of applications in various optimization problems. They typically arise as the continuum limit of exchangeable particle systems evolving by some mean-field interaction involving a gradient-type potential. However, in many problems, such as in multi-layer neural networks, the so-called particles are edge weights on large graphs whose nodes are exchangeable. Such large graphs are known to converge to continuum limits called graphons as their size grow to infinity. We show that the Euclidean gradient flow of a suitable function of the edge-weights converges to a novel continuum limit given by a curve on the space of graphons that can be appropriately described as a gradient flow or, more technically, a curve of maximal slope. Several natural functions on graphons, such as homomorphism functions and the scalar entropy, are covered by our set-up, and the examples have been worked out in detail.

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。

↑