lbfgs: Ecient L-BFGS and OWL-QN Optimization in R
lbfgs: Ecient L-BFGS and OWL-QN Optimization in R
复制标题
DOI：
--
复制
发表时间：
2013
影响因子：
3.1
通讯作者：
A. Coppola;Brandon M Stewart
中科院分区：
文献类型：
--
作者：
A. Coppola;Brandon M Stewart
本小节介绍了R的lbfgs包，它由围绕由Naoaki Okazaki编写的libLBFGS优化库构建的包装器组成。LBFGS程序包实现了有限内存Broyden-Fletcher-Goldfarb-Shanno(L-BFGS)和Orthant-Wise有限内存准牛顿(OWL-QN)优化算法。L-BFGS算法通过迭代计算逆海森矩阵的逼近，解决了给定目标梯度的最小化问题。OWL-QN算法求出目标的最优解加上问题参数的L1范数。该程序包提供了这些优化例程的快速和特定于内存的实现，特别适用于高维问题。在微基准测试中，lbfgs包可以与其他针对R的优化包相媲美。
This vignette introduces the lbfgs package for R, which consists of a wrapper built around the libLBFGS optimization library written by Naoaki Okazaki. The lbfgs package implements both the Limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) and the Orthant-Wise Limited-memory Quasi-Newton (OWL-QN) optimization algorithms. The L-BFGS algorithm solves the problem of minimizing an objective, given its gradient, by iteratively computing approximations of the inverse Hessian matrix. The OWL-QN algorithm nds the optimum of an objective plus the L1 norm of the problem’s parameters. The package oers a fast and memory-ecient implementation of these optimization routines, which is particularly suited for high-dimensional problems. The lbfgs package compares favorably with other optimization packages for R in microbenchmark tests.