arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2510.03809stat.MLcs.LG
收藏

谱阈值与可识别性与稳定性:高维学习中的有限样本相变

Spectral Thresholds for Identifiability and Stability:Finite-Sample Phase Transitions in High-Dimensional Learning

Taiwan Semiconductor Manufacturing Company (TSMC)(台湾半导体制造公司)

William Hao-Cheng Huang

AI总结:

本文提出Fisher阈值定理,通过谱正则化方法在高维学习中实现有限样本下的稳定性与可识别性,揭示了样本量与维度比的尖锐相变特性。

摘要:

在高维学习中,模型在样本量低于临界水平时会突然崩溃,这种不稳定性并非特定于算法,而是一种几何机制:当最弱的Fisher特征方向低于样本波动时,可识别性会失败。我们的Fisher阈值定理通过证明稳定性要求最小Fisher特征值超过显式的O(√(d/n))界限来正式化这一点。与以往的渐近或模型特定标准不同,这个阈值是有限样本且必要的,标志着可靠集中与不可避免失败之间的尖锐相变。为了使原理具有建设性,我们引入了Fisher地板,一种可验证的谱正则化,对平滑和预处理具有鲁棒性。在高斯混合和逻辑模型上的合成实验验证了预测的相变,与d/n缩放一致。统计上,该阈值将经典特征值条件尖锐化为非渐近定律;学习理论上,它定义了谱样本复杂性前沿,连接理论与稳健高维推断的诊断。

英文摘要:

In high-dimensional learning, models remain stable until they collapse abruptly once the sample size falls below a critical level. This instability is not algorithm-specific but a geometric mechanism: when the weakest Fisher eigendirection falls beneath sample-level fluctuations, identifiability fails. Our Fisher Threshold Theorem formalizes this by proving that stability requires the minimal Fisher eigenvalue to exceed an explicit $O(\sqrt{d/n})$ bound. Unlike prior asymptotic or model-specific criteria, this threshold is finite-sample and necessary, marking a sharp phase transition between reliable concentration and inevitable failure. To make the principle constructive, we introduce the Fisher floor, a verifiable spectral regularization robust to smoothing and preconditioning. Synthetic experiments on Gaussian mixtures and logistic models confirm the predicted transition, consistent with $d/n$ scaling. Statistically, the threshold sharpens classical eigenvalue conditions into a non-asymptotic law; learning-theoretically, it defines a spectral sample-complexity frontier, bridging theory with diagnostics for robust high-dimensional inference.

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。

↑