We study a localized notion of uniform convergence known as an "optimistic rate" (Panchenko 2002; Srebro et al. 2010) for linear regression with Gaussian data. Our refined analysis avoids the hidden constant and logarithmic factor in existing results, which are known to be crucial in high-dimensional settings, especially for understanding interpolation learning. As a special case, our analysis recovers the guarantee from Koehler et al. (2021), which tightly characterizes the population risk of low-norm interpolators under the benign overfitting conditions. Our optimistic rate bound, though, also analyzes predictors with arbitrary training error. This allows us to recover some classical statistical guarantees for ridge and LASSO regression under random designs, and helps us obtain a precise understanding of the excess risk of near-interpolators in the over-parameterized regime.
翻译:我们研究了一种地方性的统一趋同概念,称为“乐观率”(Panchenko,2002年;Srebro等人,2010年),用Gaussian数据进行线性回归。我们经过精细的分析避免了现有结果中隐藏的常数和对数因素,在高维环境中,这些常数和对数因素是众所周知的,对于理解内插学学习至关重要。作为一个特例,我们的分析恢复了Koehler等人的担保(2021年),它紧紧地体现了在无害的过度适应条件下低中低中层间插头的人口风险。但我们的乐观率约束也分析了带有任意训练错误的预测器。这使我们能够在随机设计下恢复对脊脊和LASSO回归的一些典型的统计保证,帮助我们准确了解过度分离制度下近端内插头的过度风险。