← 研踪 PaperTrailarXiv:2611.19738 · v1
Relax-and-Sparsify Approximation Algorithms for Sparse Ridge Regression
首次提交 2026-11-15 10:49
版本更新 2026-11-15 10:49
雷达观测 2026-11-17 12:06
摘要
This paper studies sparse ridge regression in the low-rank, high-dimensional setting that often arises when data collection is expensive. By exploiting the low-rank structure, we develop approximation algorithms within a relax-and-sparsify framework: solve a convex relaxation and then sparsify its solution to satisfy the original sparsity constraint. Specifically, we establish a tight rank-dependent sparsity bound for the convex relaxation by decomposing its support into two parts. This decomposition in turn motivates two approximation strategies, parameter reduction and binary redistribution, each exploiting a different part. Experiments on synthetic and real datasets show that the binary redistribution algorithms outperform the benchmarks in most instances.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-15 10:49 · 最新观测版本
Relax-and-Sparsify Approximation Algorithms for Sparse Ridge Regression
This paper studies sparse ridge regression in the low-rank, high-dimensional setting that often arises when data collection is expensive. By exploiting the low-rank structure, we develop approximation algorithms within a relax-and-sparsify framework: solve a convex relaxation and then sparsify its solution to satisfy the original sparsity constraint. Specifically, we establish a tight rank-dependent sparsity bound for the convex relaxation by decomposing its support into two parts. This decomposition in turn motivates two approximation strategies, parameter reduction and binary redistribution, each exploiting a different part. Experiments on synthetic and real datasets show that the binary redistribution algorithms outperform the benchmarks in most instances.
查看此版本解读 →查看此版本原文 ↗