arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.06434cs.LGmath.OC
收藏

基于近端牛顿法的高效图拉普拉斯估计

Efficient Graph Laplacian Estimation by Proximal Newton

Ben-Gurion University(本-古里安大学)

Yakov Medvedovsky, Eran Treister, Tirza Routtenberg

AI总结:

针对拉普拉斯约束高斯马尔可夫随机场的图学习问题,采用非凸MCP惩罚替代ℓ₁范数以降低偏差,提出基于共轭梯度、预处理等技术的二阶近端牛顿求解器,在计算效率和学习准确率上均优于现有方法。

摘要:

拉普拉斯约束高斯马尔可夫随机场(LGMRF)是从给定数据中学习加权稀疏依赖图的常用多元统计模型。该图学习问题可表述为精度矩阵的最大似然估计(MLE),需满足拉普拉斯结构约束,并带有稀疏诱导惩罚项。本文旨在准确且高效地解决这一学习问题。首先,由于常用的$\boldsymbol{\u2113}_1$-范数惩罚不适用于该场景且可能导致生成完全图,我们采用非凸的最小最大凹惩罚(MCP),该惩罚能生成估计偏差更低的稀疏解。其次,与现有针对该问题的一阶方法不同,我们开发了一种二阶近端牛顿方法以获得高效求解器,该方法利用了多项算法特性,例如使用共轭梯度法、预处理以及主动集/自由集拆分。数值实验表明,与现有方法相比,所提方法在计算复杂度和图学习准确率两方面均具备优势。

英文摘要:

The Laplacian-constrained Gaussian Markov Random Field (LGMRF) is a common multivariate statistical model for learning a weighted sparse dependency graph from given data. This graph learning problem can be formulated as a maximum likelihood estimation (MLE) of the precision matrix, subject to Laplacian structural constraints, with a sparsity-inducing penalty term. This paper aims to solve this learning problem accurately and efficiently. First, since the commonly used $\ell_1$-norm penalty is inappropriate in this setting and may lead to a complete graph, we employ the nonconvex minimax concave penalty (MCP), which promotes sparse solutions with lower estimation bias. Second, as opposed to existing first-order methods for this problem, we develop a second-order proximal Newton approach to obtain an efficient solver, utilizing several algorithmic features, such as using Conjugate Gradients, preconditioning, and splitting to active/free sets. Numerical experiments demonstrate the advantages of the proposed method in terms of both computational complexity and graph learning accuracy compared to existing 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。

↑