arXiv:2502.18213cs.DScs.LG2025-02ICLR被引 3

提出首个近似最优的单指标模型主动回归算法,大幅减少对目标向量的查询次数。

Near-optimal Active Regression of Single-Index Models

  • 设计新算法,在 Lipschitz 条件下实现 (1+ε) 近似解。
  • 查询次数为 Õ(d^{p/2∨1}/ε^{p∨2}),对 p∈[1,2] 和 p>2 均接近最优。
  • 适用于需高效获取目标数据的高维优化场景,如稀疏感知或实验设计。

单指标模型的主动回归问题旨在求解最小化‖f(Ax)−b‖_p,其中矩阵A完全可访问,而向量b只能通过逐项查询获取,目标是尽可能减少对b的查询次数。当f满足Lipschitz条件时,以往方法仅能获得常数因子近似。本文首次提出一个(1+ε)-近似算法,仅需查询Õ(d^{p/2∨1}/ε^{p∨2})个b的分量。该查询复杂度对p∈[1,2]被证明在对数因子内最优,且当p>2时,1/ε^p的ε依赖性也被证明最优。

原文摘要 · Abstract (English)

The active regression problem of the single-index model is to solve $\min_x \lVert f(Ax)-b\rVert_p$, where $A$ is fully accessible and $b$ can only be accessed via entry queries, with the goal of minimizing the number of queries to the entries of $b$. When $f$ is Lipschitz, previous results only obtain constant-factor approximations. This work presents the first algorithm that provides a $(1+\varepsilon)$-approximation solution by querying $\tilde{O}(d^{\frac{p}{2}\vee 1}/\varepsilon^{p\vee 2})$ entries of $b$. This query complexity is also shown to be optimal up to logarithmic factors for $p\in [1,2]$ and the $\varepsilon$-dependence of $1/\varepsilon^p$ is shown to be optimal for $p>2$.

主动学习回归分析优化算法

Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。