TL;DR该研究提出了一种贪心算法,Gradient Support Pursuit (GraSP),以近似任意形式损失函数的稀疏极小值,适用于稀疏逻辑回归等问题,算法性能通过在合成数据上的数值模拟进行评估。
Abstract
sparsity-constrained optimization has wide applicability in machine learning,
statistics, and signal processing problems such as feature selection and
compressive Sensing. A vast body of work has studied the sparsity-constrained
optimization from theoretical, algorithmic, and applicati